This is the accessible text file for GAO report number GAO-04-801 
entitled 'Pipeline Safety: Management of the Office of Pipeline 
Safety's Enforcement Program Needs Further Strengthening' which was 
released on July 23, 2004.

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as part 
of a longer term project to improve GAO products' accessibility. Every 
attempt has been made to maintain the structural and data integrity of 
the original printed product. Accessibility features, such as text 
descriptions of tables, consecutively numbered footnotes placed at the 
end of the file, and the text of agency comment letters, are provided 
but may not exactly duplicate the presentation or format of the printed 
version. The portable document format (PDF) file is an exact electronic 
replica of the printed version. We welcome your feedback. Please E-mail 
your comments regarding the contents or accessibility features of this 
document to Webmaster@gao.gov.

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately.

Report to Congressional Recipients: 

United States Government Accountability Office: 

July 2004: 

PIPELINE SAFETY: 

Management of the Office of Pipeline Safety's Enforcement Program Needs 
Further Strengthening: 

GAO-04-801: 

GAO Highlights: 

Highlights of GAO-04-801, a report to congressional recipients 

Why GAO Did This Study: 

While pipelines are inherently safer to the public than other modes of 
freight transportation, pipeline accidents involving natural gas and 
hazardous liquids (such as gasoline) can have serious consequences. 
For example, a natural gas pipeline ruptured near Carlsbad, New Mexico, 
in 2000, killed 12 people, and resulted in $1 million in damages or 
losses. The Office of Pipeline Safety (OPS) administers the national 
regulatory program to ensure safe pipeline transportation. OPS uses 
its enforcement program, when safety problems are found, as one means 
to do so.

This study reports on (1) the effectiveness of OPS’s enforcement 
strategy and (2) OPS’s actions for assessing monetary sanctions (civil 
penalties), among other things.

What GAO Found: 

The effectiveness of OPS’s enforcement strategy cannot be determined 
because the agency has not set goals for its enforcement program, fully 
defined its strategy, or established performance measures linked to 
goals that would allow an assessment of results. (See below.) These 
are key elements of effective management. Without these elements, the 
agency cannot determine whether recent changes in its strategy are 
having the desired effects on pipeline safety. Over the past several 
years, OPS has placed priority on other areas—developing a new 
risk-based regulatory approach—and it believes these efforts will 
change the safety culture of the industry. OPS now intends to devote 
more attention to strengthening the management of the agency’s 
enforcement program. In particular, OPS is developing an enforcement 
policy that will help define its enforcement strategy and has made some 
initial steps toward identifying new performance measures. However, 
OPS does not anticipate finalizing such a policy until sometime during 
2005 and lacks a systematic approach for incorporating some of the key 
practices identified for achieving successful performance measurement 
systems. 

[See PDF for image]

[End of figure]

OPS has increased both the number and the size of the penalties it has 
assessed against pipeline operators over the last 4 years (2000 through 
2003) following its decision to be “tough but fair” in assessing 
penalties. During this period, OPS assessed an average of 22 penalties 
per year, compared with an average of 14 per year for the previous 5 
years (1995 through 1999), a period of more lenient enforcement. In 
addition, the average penalty amount increased from $18,000 to $29,000 
over the two periods. While civil penalty use and size has increased, 
it is not clear whether this action will help deter noncompliance with 
the agency’s safety regulations. Stakeholders expressed differing 
views: some thought that any penalty had a deterrent effect if it kept 
the pipeline operator in the public eye, while others told us that the 
penalties were too small to be effective sanctions. About 94 percent of 
the 216 penalties levied from 1994 through 2003 have been paid. 
However, OPS lacks effective management controls to assure that 
penalties are collected. For example, OPS does not routinely inform 
its collection agent of penalties it has assessed. 

What GAO Recommends: 

GAO recommends that OPS define its enforcement goals and strategy and 
establish a systematic approach for designing new performance measures. 
GAO also makes several recommendations aimed at improving management 
control over the collection of civil penalties.

GAO provided a draft of this report to the Department of Transportation 
for its review and comment. The department generally agreed with the 
information in the report and its recommendations.

www.gao.gov/cgi-bin/getrpt?GAO-04-801.

To view the full product, including the scope and methodology, click on 
the link above. For more information, contact Katherine Siggerud, (202) 
512-2834, siggerudk@gao.gov.

[End of section]

Contents: 

Letter: 

Results in Brief:  

Background:  

Key Management Elements Are Needed to Determine the Effectiveness of 
OPS's Enforcement Strategy:  

OPS Has Increased Its Use of Civil Penalties; OPS and FAA Need Stronger 
Management Controls Over Collections:  

Conclusions:  

Recommendations for Executive Action:  

Agency Comments and Our Evaluation:  

Appendixes: 

Appendix I: Scope and Methodology: 

Other Aspects of Our Work:  

Organizations Contacted:  

Appendix II: Other Pipeline Safety Enforcement Issues: 

Appendix III: OPS's Implementation of Recommendations Made in Our May 
2000 Report: 

Appendix IV: OPS's Enforcement Policies Influenced Trends in 
Enforcement and Administrative Actions:  

Appendix V: Contacts and Staff Acknowledgments: 

Contacts:  

Staff Acknowledgments:  

Tables: 

Table 1: OPS's Enforcement and Administrative Actions: 

Table 2: Enforcement Program Performance Measures that OPS Currently 
Uses and Is Considering Developing:  

Table 3: OPS's Proposed and Assessed Civil Penalties, 1994 through 
2003: 

Table 4: Reductions in OPS's Proposed Civil Penalties, 1994 through 
2003: 

Table 5: OPS's and FAA's Regulations and Processes for Issuing Civil 
Penalties:  

Table 6: OPS's and FAA's Use of Civil Penalties, 1994 through 2002:  

Figures: 

Figure 1: OPS's Use of Civil Penalties, 2000 through 2003 Compared with 
1995 through 1999: 

Figure 2: Number of Accidents and Accident Rates for Interstate 
Hazardous Liquid Pipelines, 1994 through 2003: 

Figure 3: Number of Accidents on Interstate Natural Gas Pipelines, 1994 
through 2003: 

Figure 4: OPS's Use of Selected Enforcement Actions in 1999 and 2003: 

Figure 5: Example of How OPS's Enforcement Strategy Contributes to 
Pipeline Safety: 

Figure 6: Operators' Payment of Civil Penalties, 1994 through 2003:  

Figure 7: OPS's Policy Eras and Enforcement and Administrative Actions, 
1994 through 2003: 

Figure 8: OPS's Enforcement and Administrative Actions and Interstate 
Pipeline Accidents per 10,000 Miles of Pipeline, 1994 through 2003: 

Figure 9: OPS's Enforcement and Administrative Actions and Serious 
Interstate Pipeline Accidents, 1994 through 2003: 

Figure 10: OPS's Enforcement and Administrative Actions and Miles of 
Pipeline, 1994 through 2003: 

Figure 11: OPS's Enforcement and Administrative Actions and Number of 
OPS Inspectors, 1994 through 2003: 

Figure 12: OPS's Enforcement and Administrative Actions and Inspections 
Conducted, 1994 through 2003: 

Figure 13: OPS's Enforcement and Administrative Actions and Time Taken 
to Conduct Inspections, 1994 through 2003: 

Figure 14: OPS's Enforcement and Administrative Actions and Inspections 
Conducted, by Type of Inspection, 1994 through 2003: 

Figure 15: OPS's Enforcement and Administrative Actions and Natural Gas 
Consumption, 1994 through 2003: 

Figure 16: OPS's Enforcement and Administrative Actions and Petroleum 
Consumption, 1994 through 2003: 

Figure 17: OPS's Enforcement and Administrative Actions and New 
Privately Owned Housing Units Started, 1994 through 2003: 

Figure 18: OPS's Enforcement and Administrative Actions and the Total 
Value of Construction Put in Place, 1994 through 2003: 

Abbreviations:

FAA: Federal Aviation Administration: 

OPS: Office of Pipeline Safety: 

RSPA: Research and Special Programs Administration: 

Letter July 23, 2004: 

Congressional Recipients: 

While pipelines are inherently safer to the public than other modes of 
freight transportation, the volatile nature of the products they carry 
means that pipeline accidents can have serious consequences. For 
example, when a natural gas pipeline ruptured near Carlsbad, New 
Mexico, in 2000, the released gas ignited and killed 12 people. 
Property and other damages or losses totaled almost $1 million.

The Office of Pipeline Safety (OPS), within the Department of 
Transportation's Research and Special Programs Administration (RSPA), 
administers the national regulatory program to ensure the safe 
transportation of natural gas and hazardous liquids by 
pipeline.[Footnote 1] The office attempts to ensure the safe operation 
of pipelines through regulation, national consensus 
standards,[Footnote 2] and oversight of the industry through 
inspections and enforcement actions when safety problems are found. The 
office uses a variety of enforcement tools, including some, such as 
compliance orders, that require pipeline operators to correct 
underlying safety violations. OPS may also impose monetary sanctions 
(civil penalties) on pipeline operators for violations of its pipeline 
safety regulations.

In May 2000, we reported, among other things, that OPS had reduced its 
use of civil penalties and increased its use of administrative actions 
(such as warning letters) without assessing the effects of these 
changes on pipeline safety compliance.[Footnote 3] The Pipeline Safety 
Improvement Act of 2002 directed that we (1) evaluate the effectiveness 
of OPS's enforcement strategy and (2) examine OPS's assessment of civil 
penalties. These topics form the major focus of this report. To meet 
other requirements of the act and your interests, we are also reporting 
on how OPS enforces pipeline safety and how its enforcement policies 
and procedures have changed over time, whether OPS substitutes civil 
penalties for another form of enforcement, and how OPS's use of civil 
penalties compares with that of the Federal Aviation Administration 
(FAA), which oversees aviation safety. These issues are discussed in 
appendix II. Finally, we are reporting on how the office responded to 
the recommendations in our May 2000 report and whether industry and 
economic trends have influenced OPS's enforcement actions. These issues 
are discussed in appendixes III and IV.

To evaluate the effectiveness of OPS's enforcement strategy, we 
determined the extent to which the office's strategy incorporates 
several elements that are important for effective program management: 
clear program goals, a well-defined strategy for achieving goals, and 
performance measures that are linked to program goals. We identified 
elements of effective program management by reviewing our products on 
this topic as well as other key literature, such as Office of 
Management and Budget guidance, on this subject. We interviewed OPS 
officials and reviewed documents about the agency's activities and 
plans related to each of these elements. We interviewed pipeline safety 
stakeholders, including industry trade associations, federal agencies, 
state pipeline agencies and associations, a local representative from 
Virginia, and pipeline safety advocacy groups, to determine their views 
on the effectiveness of OPS's enforcement efforts. To examine OPS's 
assessment of civil penalties against pipeline operators, we analyzed 
the extent to which OPS proposed and assessed civil penalties from 1994 
through 2003 and pipeline operators have paid them.[Footnote 4] (FAA's 
general accounting division collects civil penalties for OPS.) We also 
analyzed data from OPS on its enforcement actions and from FAA on its 
collection of civil penalties. We examined OPS's procedures and actions 
to ensure that the safety violations that lead to civil penalties are 
remedied. We interviewed stakeholders on whether OPS's civil penalties 
help deter safety violations. As part of our review, we assessed the 
internal controls and the reliability of the data elements needed for 
this engagement. We determined that the data elements were sufficiently 
reliable for our purposes, except for those that would allow us to 
determine the timeliness of civil penalty payments. This exception did 
not create a major impediment in reporting on OPS's use of civil 
penalties. We performed our review in accordance with generally 
accepted government auditing standards from June 2003 through July 
2004. (See app. I for additional details on our scope and 
methodology.)

Results in Brief: 

Although in recent years OPS has made a number of changes in its 
enforcement strategy that have the potential to improve pipeline 
safety, the effectiveness of this strategy cannot currently be 
determined because the agency has not incorporated three key elements 
of effective program management--clear program goals, a well-defined 
strategy for achieving those goals, and performance measures linked to 
the program goals. Without these three key elements, OPS cannot 
determine whether recent and planned changes in its enforcement 
strategy are having or will have the desired effects on pipeline 
safety. Under a more aggressive "tough but fair" enforcement strategy 
adopted in 2000, OPS has increased its use of civil penalties and other 
enforcement tools. However, OPS has not established goals that specify 
the intended results of this strategy, developed a policy that 
describes the strategy and its contribution to pipeline safety, or put 
measures in place that would allow OPS to determine and demonstrate the 
effects of this new strategy on pipeline safety. OPS officials told us 
that over the past several years they have placed priority on other 
areas--such as developing a new risk-based regulatory approach that is 
focused on ensuring that pipeline operators find and fix threats to 
pipeline safety in areas where an accident could have the most serious 
consequences--and that they believe these efforts will change the 
safety culture of the industry and have a greater impact on safety than 
enforcement. According to OPS officials, they have made significant 
progress in implementing this new approach and now intend to devote 
more attention to strengthening the management of the agency's 
enforcement program. OPS is working on developing an enforcement policy 
that will help define its enforcement strategy and has taken some 
initial steps toward identifying performance measures that could 
provide better information for managing enforcement efforts. However, 
the agency does not anticipate finalizing this policy until sometime in 
2005. Additionally, OPS does not have a systematic approach in place 
that incorporates key practices for achieving successful performance 
measurement systems, such as linking performance measures to program 
goals.

OPS increased both the number and the size of its civil penalties in 
response to criticism that its enforcement activities were weak and 
ineffective. Specifically, in the last 4 years (2000 through 2003), 
after it decided to be "tough but fair" in assessing civil penalties, 
OPS assessed an average 22 penalties per year, compared with an average 
of 14 per year during the previous 5 years (1995 through 1999), when 
OPS's policy was to "partner" with industry, rather than primarily to 
enforce compliance. In addition, the average civil penalty that OPS 
assessed from 2000 through 2003 was about $29,000, compared with an 
average penalty of about $18,000 during the previous 5 years. (See fig. 
1.) OPS assessed the penalty that it proposed 69 percent of the time 
(150 of 216 civil penalties). For the remaining 66 penalties, OPS 
reduced the assessments by about 37 percent--from a total of about $2.8 
million to about $1.7 million. OPS's database does not provide summary 
information on why penalties are reduced. As a result, we are not able 
to provide information on the most common reasons why penalties were 
reduced. OPS does not use civil penalties as extensively as other 
enforcement actions; they represent about 14 percent of all the 
enforcement and administrative actions OPS has taken over the past 10 
years.

Figure 1: OPS's Use of Civil Penalties, 2000 through 2003 Compared with 
1995 through 1999: 

[See PDF for image]

Note: This figure does not include the $250,000 civil penalty OPS 
assessed Shell Pipeline Company for the June 1999 pipeline accident in 
Bellingham, Washington. See discussion later in this report.

[End of figure]

While OPS has increased both the number and the size of its civil 
penalties, it is not clear whether this action will help deter 
noncompliance with the agency's safety regulations. The pipeline safety 
stakeholders we spoke with expressed differing views on this issue. 
Some--such as pipeline industry officials--said that civil penalties of 
any size act as a deterrent, in part because they keep companies in the 
public eye. Others--such as pipeline safety advocacy groups--said that 
OPS's civil penalties are too small to deter noncompliance. Finally, 
departmental data show that operators have paid 202 of the 216 civil 
penalties (94 percent) over the past 10 years. However, OPS and, to a 
lesser degree, FAA (whose general accounting division collects civil 
penalties for OPS) lack important management controls to ensure that 
penalties are collected. For example, although most civil penalties are 
paid, FAA is not aware of all the penalties that it may ultimately be 
responsible for collecting because OPS does not routinely notify FAA of 
proposed or assessed civil penalties. We found that FAA had no record 
of 44 of the 290 civil penalties that OPS proposed from 1994 through 
2003.

We are making several recommendations to improve OPS's ability to 
determine and demonstrate the effectiveness of its enforcement strategy 
and to make adjustments in this strategy as needed. We are also making 
recommendations to improve OPS's and FAA's management controls over the 
collection of civil penalties.

We provided a draft of this report to the Department of Transportation 
for its review and comment. Departmental representatives generally 
agreed with the information in the report and its recommendations.

Background: 

Pipelines transport the bulk of natural gas and hazardous liquids in 
the United States. Specifically, pipelines carry nearly all of the 
natural gas and about two-thirds of the crude oil and refined oil 
products. Three primary types of pipelines form a 2.4-million-mile 
network across the United States.

* Natural gas transmission pipelines transport natural gas over long 
distances from sources to communities (about 327,000 miles, primarily 
interstate).

* Natural gas distribution pipelines continue to transport natural gas 
from transmission lines to consumers (about 1.9 million miles, 
primarily intrastate).

* Hazardous liquid pipelines transport products, such as crude oil, to 
refineries and the refined product on to product terminals (about 
161,000 miles, primarily interstate).

Pipelines have an inherent safety advantage over other modes of freight 
transportation because they are primarily located underground, away 
from public contact. By one measure, the reduction in accidents 
overall, the hazardous liquid pipeline industry has greatly improved 
its safety record over the past 10 years. (See fig. 2.) From 1994 
through 2003, accidents on interstate hazardous liquid pipelines 
decreased from 245 in 1994 to 126 in 2003, or almost 49 percent.
[Footnote 5] These accidents resulted in an average of 2 fatalities 
and 8 injuries per year. However, the industry's safety record has not 
improved for accidents with the greatest consequence--those resulting 
in a fatality, an injury, or in property damage of $50,000 or more--
called serious accidents in this report. The number of serious 
accidents stayed about the same over the 10-year period.[Footnote 6] 
The lack of significant change over time in the number of serious 
accidents on interstate hazardous liquid pipelines may be due in part 
to the relatively small number of these accidents--about 88 every year. 
The accident rate--which considers the amount of product and the 
distance it is shipped--followed a similar pattern. The accident rate 
for hazardous liquid pipelines overall decreased from about 0.41 
accidents per billion ton-miles shipped in 1994 to about 0.25 accidents 
per billion ton-miles shipped in 2002.[Footnote 7] The accident rate 
for serious interstate hazardous liquid pipeline accidents stayed the 
same, averaging about 0.15 accidents per billion ton-miles shipped from 
1994 through 2002.

Figure 2: Number of Accidents and Accident Rates for Interstate 
Hazardous Liquid Pipelines, 1994 through 2003: 

[See PDF for image] 

Notes: The hazardous liquid accident rate is expressed in terms of 
accidents per billion ton-miles of petroleum products shipped. Federal 
agencies and industry associations we contacted could not provide data 
on other hazardous liquids shipped.

[End of figure] 

Aggregated industry data on the amounts of products shipped through 
hazardous liquid pipelines for 2003 are not available, so we do not 
present accident rate information for this year.

In contrast to hazardous liquid pipelines, accidents on interstate 
natural gas pipelines increased from 81 in 1994 to 97 in 2003, or 
almost 20 percent. (See fig. 3.) These accidents resulted in an average 
of 3 fatalities and 10 injuries per year. The number of serious 
accidents on interstate natural gas pipelines also increased, from 64 
in 1994 to 84 in 2003, though they have fluctuated considerably over 
this period. Information on accident rates for natural gas pipelines is 
not available because of the lack of data on natural gas shipped 
through pipelines. As with hazardous liquid pipelines, the lack of 
significant change over time in the number of total accidents and 
serious accidents on interstate natural gas pipelines may be due in 
part to the relatively small number of these accidents--about 65 every 
year.

Figure 3: Number of Accidents on Interstate Natural Gas Pipelines, 1994 
through 2003: 

[See PDF for image] 

Note: Data on natural gas shipped through interstate pipelines are not 
available; these data are needed to calculate the accident rate for 
this type of pipeline.

[End of figure] 

OPS, within RSPA, administers the national regulatory program to ensure 
the safe operation of the nation's natural gas and hazardous liquid 
pipelines. OPS has carried out its oversight responsibility by 
developing and issuing prescriptive minimum safety standards and 
enforcing these standards. Recently, the agency has developed 
additional standards that are risk based and focus on improving 
pipeline operators' management of their operations rather than on 
meeting prescriptive requirements. In 1999, to reduce the risk of 
accidents attributable to human error, OPS issued a new operator 
qualification regulation requiring pipeline operators to develop 
programs for ensuring that individuals working on their pipeline 
systems are qualified to do so. In 2000, to better focus on safety 
risks that are unique to individual pipelines, OPS issued the first in 
a series of integrity management regulations requiring operators to 
better protect pipeline segments where a leak or rupture could have a 
significant effect on densely populated or environmentally sensitive 
areas (called high-consequence areas).[Footnote 8] Under this new risk-
based regulatory approach, operators must, in addition to meeting 
minimum safety requirements, develop comprehensive plans for 
identifying the range of risks facing these segments and taking actions 
to mitigate these risks. According to OPS, it is devoting a large 
portion of its resources to implementing the integrity management 
program. OPS issued integrity management requirements for large 
hazardous liquid pipeline operators (those with 500 or more miles of 
pipeline) in December 2000, for small hazardous liquid pipeline 
operators in January 2002, and for natural gas transmission pipeline 
operators in December 2003. The agency is carrying out inspections of 
operators' compliance with these requirements in separate phases, 
starting with inspections of large hazardous liquid operators from 
September 2002 through April 2004. In all, the agency will need to 
inspect the integrity management programs of more than 1,000 individual 
operators of hazardous liquid and natural gas transmission pipelines.

To improve pipeline safety, OPS carries out several types of 
activities. First, it develops and issues pipeline safety regulations 
and supports national consensus standards, which provide additional 
guidance to pipeline operators in managing their pipeline systems 
safely. In addition, OPS undertakes oversight activities, which include 
inspections to determine compliance with its regulations, accident 
investigations, and enforcement. Finally, OPS administers other 
programs--including, for example, research and development to enhance 
pipeline safety technologies, data collection to better define 
pipeline-related problems and concerns, and education to prevent 
excavation-related damage.

When OPS finds a violation--such as the failure of an operator to 
inspect various aspects of its pipeline--during an inspection or an 
investigation after an accident, it may take one of several types of 
enforcement or administrative actions depending on the nature and 
severity of the violation. (See table 1.) An enforcement action may 
require the operator to correct an unsafe condition or practice, or the 
enforcement action may be a civil penalty (monetary fine). An 
administrative action notifies a pipeline operator of a safety concern 
that is not serious enough to require an enforcement action.

Table 1: OPS's Enforcement and Administrative Actions: 

Enforcement action: Corrective action order; 
Description: Notifies an operator to remedy pipeline facility 
operations that OPS considers hazardous to life or property.

Enforcement action: Civil penalty; 
Description: Imposes a monetary penalty on an operator that violates a 
safety rule.

Enforcement action: Compliance order; 
Description: Notifies an operator of actions required to bring 
operations and facilities into compliance with safety rules.

Enforcement action: Notice of amendment; 
Description: Tells an operator to amend existing procedures to correct 
deficiencies to ensure the safe operation of a pipeline facility.

Enforcement action: Safety order (under development); 
Description: Will order an operator to remedy safety-related conditions 
that could significantly change or restrict pipeline operations (these 
conditions do not pose a threat to life or property).

Administrative action: Warning letter; 
Description: Notifies an operator of probable violations of pipeline 
safety rules, but the probable violation does not cause an immediate 
hazard or irreparable damage to the facility.

Administrative action: Letter of concern; 
Description: Notifies an operator of a minor safety concern and of the 
action to be taken to address the problem. 

Source: GAO presentation of OPS information.

[End of table]

When imposing civil penalties, OPS must by law consider seven factors: 
(1) the nature, circumstances, and gravity of the violation; (2) the 
degree of the operator's culpability; (3) the operator's history of 
prior offenses; (4) the operator's ability to pay; (5) any good faith 
shown by the operator in attempting to achieve compliance; (6) the 
effect on the operator's ability to continue doing business; and (7) 
other matters as justice may require. Before OPS imposes a civil 
penalty, it issues the pipeline operator a notice of probable violation 
that documents the alleged violation and identifies the proposed civil 
penalty amount. OPS then allows the operator to present additional 
evidence either in writing or in an informal hearing. Attorneys from 
RSPA's Office of Chief Counsel preside over these hearings. Evidence 
presented by the operator may result in the civil penalty being 
affirmed, reduced, or withdrawn. If, after this step, the hearing 
officer determines that a violation occurred, OPS's associate 
administrator issues a final order that requires the operator to 
correct the safety violation (if needed) and pay the penalty (termed 
"assessed penalties" in this report).[Footnote 9] The operator has 20 
days after the final order is received to pay the penalty. FAA collects 
civil penalties for OPS.[Footnote 10]

From 1992 through 2002, federal law allowed OPS to assess up to $25,000 
for each day that a violation continued, not to exceed $500,000 for any 
related series of violations. In December 2002, the Pipeline Safety 
Improvement Act increased these amounts to $100,000 and $1 million, 
respectively.

OPS is a small federal agency. In fiscal year 2003, OPS employed about 
150 people--about half of whom were pipeline inspectors. In contrast, 
the Federal Railroad Administration, another agency within the 
Department of Transportation, employs 855 people, including more than 
400 inspectors to enforce rail safety regulations. In addition, FAA, 
the agency within the Department of Transportation responsible for the 
safety of civil aviation, employed about 48,500 people in fiscal year 
2003. About 4,000 of these employees were safety inspectors. For fiscal 
year 2003, OPS received about $66.8 million in appropriations and about 
$17.5 million from the Pipeline Safety Fund.

OPS retains full responsibility for enforcing regulations on interstate 
pipelines, and it certifies states to perform these functions for 
intrastate pipelines.[Footnote 11] Currently, OPS has agreements with 
11 states, known as interstate agents, to help it inspect segments of 
interstate pipeline within these states' boundaries. However, OPS 
undertakes any enforcement actions identified through inspections 
conducted by interstate agents. In 2002, about 400 state pipeline 
safety inspectors assisted OPS in overseeing pipeline safety within 
their states, according to the latest available data.

Key Management Elements Are Needed to Determine the Effectiveness of 
OPS's Enforcement Strategy: 

Although in recent years OPS has made a number of changes in its 
enforcement strategy that have the potential to improve pipeline 
safety, the effectiveness of this strategy cannot currently be 
determined because the agency has not incorporated three key elements 
of effective program management--clear program goals, a well-defined 
strategy for achieving those goals, and performance measures linked to 
the program goals.

OPS's Enforcement Strategy, as Well as a Broader Oversight Approach, 
Has Been Evolving: 

OPS's enforcement strategy, as well as its overall approach for 
overseeing pipeline safety, has undergone significant changes in the 
last 5 years. Before 2000, the agency had emphasized partnering with 
the pipeline industry to improve pipeline safety rather than punishing 
noncompliance. In 2000, in response to concerns that its enforcement 
was weak and ineffective, the agency decided to institute a "tough but 
fair" enforcement approach and committed to making greater use of all 
its enforcement tools, including larger civil penalties. In 2001, to 
further strengthen its enforcement, the agency began issuing more 
corrective action orders requiring operators to address safety problems 
that led to pipeline accidents. In 2002, OPS created an Enforcement 
Office to put more focus on enforcement and help ensure consistency in 
enforcement decisions. However, the agency has not yet filled key 
positions in this office.

OPS was making these changes in its enforcement strategy at the same 
time that it was significantly changing its overall approach for 
overseeing pipeline safety. In particular, in 2000 the agency began 
implementing its new integrity management program, which requires 
operators to systematically manage risks to their pipelines in areas 
where an accident could have the highest consequences. The agency 
believes that pipeline accidents in these high-consequence areas will 
decrease because operators are required, under this risk-based 
approach, to identify and repair significant defects in pipelines 
located in these areas. Officials have emphasized that they believe 
this program is improving the safety culture of the pipeline industry 
and has a greater potential to improve safety than enforcing OPS's 
traditional minimum safety standards. According to these officials, in 
the last several years, they have placed a priority on developing and 
implementing this risk-based regulatory approach and on developing a 
sound approach for overseeing pipeline operators' fulfillment of the 
agency's new requirements. For example, OPS has developed detailed 
protocols and guidance for inspecting operators' identification of 
risks and resulting repairs and has developed new information systems 
for tracking the status of issues identified in these inspections. OPS 
has developed a similar approach for overseeing these companies' 
fulfillment of its new requirements for ensuring that their employees 
are qualified to operate pipeline systems. According to OPS officials, 
the agency plans to use these new oversight approaches as a model for 
improving its oversight of operators' compliance with its minimum 
safety requirements. Officials have emphasized that their efforts to 
raise safety standards, inspect pipeline operators against these 
standards, investigate accidents, and take enforcement actions 
collectively represent an overall systematic approach to improving 
pipeline safety.

In 2002, OPS began to enforce its new integrity management and operator 
qualification standards, in addition to its minimum safety standards. 
For integrity management, the agency has primarily used notices of 
amendment, which require improvements in procedures rather than 
stronger enforcement actions to give pipeline operators time to learn 
how to build programs that meet OPS's complex standards. OPS has 
recently started to make greater use of civil penalties in enforcing 
these standards. The agency has also used a mix of enforcement actions 
in enforcing its operator qualification standards. OPS's use of civil 
penalties, corrective action orders, and notices of amendment was 
significantly greater in 2003 than it was in 1999, the year before OPS 
started changing its enforcement strategy. (See fig. 4.)

Figure 4: OPS's Use of Selected Enforcement Actions in 1999 and 2003: 

[See PDF for image] 

[End of figure] 

According to OPS's associate administrator, the agency has made 
significant progress in implementing its integrity management program 
and now needs to devote more attention to strengthening the management 
of its enforcement program.[Footnote 12] Consequently, OPS has recently 
begun to "reengineer" this program. Efforts under way include 
developing a new enforcement policy and guidelines, developing a 
streamlined process for handling enforcement cases, modernizing and 
integrating the agency's inspection and enforcement databases, meeting 
with stakeholders to obtain their views on how to make the enforcement 
action fit the violation, and hiring additional staff devoted to 
enforcement. Some aspects of these plans are discussed in more detail 
in the following sections.

OPS Needs Goals for Its Enforcement Program: 

Although OPS has overall performance goals, the agency has not 
established specific goals for its enforcement program. According to 
OPS officials, the agency's enforcement program is designed to achieve 
OPS's overall performance goals of (1) reducing the number of pipeline 
accidents by 5 percent annually and (2) reducing the number of spills 
of oil and other hazardous liquids from pipelines by 6 percent 
annually.[Footnote 13] A number of other agency efforts--including the 
development of new safety standards, inspections, and initiatives to 
help communities prevent damage to pipelines--are also designed to 
achieve these goals.

The above performance goals are useful agencywide safety goals because 
they identify the end outcomes, or ultimate results, that OPS seeks to 
achieve through its various efforts. However, OPS has not established 
goals for its enforcement program that identify the intermediate 
outcomes, or direct results, the enforcement program seeks to achieve. 
Intermediate outcomes show progress toward achieving end outcomes. For 
example, enforcement actions can result in improvements in pipeline 
operators' safety performance that can subsequently result in reduced 
pipeline accidents and spills. OPS managers have told us that the 
desired direct results of enforcement actions are deterring 
noncompliance with safety standards, reducing repeat violations of 
specific standards, and influencing pipeline operators' safety 
performance by requiring safety improvements to correct identified 
problems. Program outputs, such as enforcement actions, can lead to 
such intermediate outcomes, which in turn can result in the desired end 
outcomes of reduced accidents and spills.[Footnote 14] (See fig. 5.)

Figure 5: Example of How OPS's Enforcement Strategy Contributes to 
Pipeline Safety: 

[See PDF for image] 

Note: The intermediate outcomes identified in this figure are examples 
of the intended direct results of OPS's enforcement strategy and are 
based on comments by OPS officials in various interviews. OPS has not 
identified these as intermediate outcomes or goals of its enforcement 
program.

[End of figure] 

We have reported that it is a useful practice for federal programs to 
complement end outcome goals with intermediate outcome goals in order 
to help show a program's contribution to desired end outcomes.[Footnote 
15] OPS is considering establishing a goal to reduce the amount of time 
it takes to issue final enforcement actions. While such a goal could be 
useful for improving the management of the enforcement program, it does 
not reflect the direct results the agency hopes to achieve through 
enforcement. Clear goals for the enforcement program that specify 
intended intermediate outcomes (such as a reduced number of repeat 
offenders) would be useful to OPS and to external stakeholders to show 
how enforcement efforts contribute to pipeline safety.

OPS Needs to Fully Define Its Enforcement Strategy: 

OPS has not fully defined its strategy for using enforcement to achieve 
its goals. According to OPS officials, the agency's increased use of 
civil penalties and corrective action orders reflects a major change in 
its enforcement strategy. However, although OPS began to implement 
these changes in 2000, it has not yet developed a policy that describes 
this new, more aggressive, enforcement strategy or how the strategy 
will contribute to the achievement of OPS's performance goals. In 
addition, although OPS's authorizing statutes and regulations provide 
general guidance on the use of various types of enforcement actions, 
the agency does not have up-to-date detailed internal guidelines on the 
use of its enforcement actions that reflect its current strategy. For 
example, OPS has an enforcement manual that provides general guidance 
on the various types of enforcement actions and how each should be 
used, but this guidance reflects the agency's earlier, more lenient, 
approach to enforcement and does not specify the types of situations 
that may warrant certain types of actions. In addition, although OPS 
began enforcing its integrity management standards and received greater 
enforcement authority under the Pipeline Safety Improvement Act in 
2002, it does not yet have guidelines in place for enforcing these 
standards or implementing the new authority provided by the 
act.[Footnote 16]

An important internal control practice is to have policies and 
procedures for each agency activity.[Footnote 17] According to agency 
officials, OPS management has communicated enforcement priorities and 
ensured consistency in enforcement decisions through frequent internal 
meetings and detailed inspection protocols and guidance. However, 
without enforcement policies and guidelines in place that reflect its 
current strategy, the agency lacks reasonable assurance that this 
strategy is being carried out effectively. For example, OPS regional 
and state inspector staff may not be fully aware of the agency's 
current strategy, and regional directors may be less likely to make 
complex judgments about enforcement in a uniform manner.

Agency officials recognize the need to develop an enforcement policy 
and up-to-date detailed enforcement guidelines and have been working on 
various aspects of this task. According to OPS officials, the agency 
has been in a period of "recreating" its enforcement policy. To date, 
the agency has completed an initial set of enforcement guidelines for 
its operator qualification standards and has developed various other 
draft guidelines. According to OPS officials, the policy and remaining 
guidelines, when completed, will: 

* reflect the agency's increased emphasis on civil penalties and 
corrective action orders and provide detailed guidance on the use of 
these and other enforcement tools;

* cover the enforcement of OPS's traditional safety standards, as well 
as its new integrity management standards; and: 

* discuss how OPS will implement the greater enforcement authority 
provided to it by the Pipeline Safety Improvement Act of 2002.

Agency officials anticipate that the new enforcement policy and 
remaining guidelines will not be finalized until sometime in 2005 
because of the complexity of these tasks.

While the development of an enforcement policy and guidelines should 
help to define OPS's enforcement strategy, it is not clear whether this 
effort will link this strategy with results, since agency officials 
have not established goals specifically for their enforcement efforts. 
We have reported on the importance for effective program management of 
connecting strategies to desired results by clearly defining program 
strategies and developing and presenting a rationale for how these 
strategies contribute to the achievement of goals.[Footnote 18]

OPS Needs Adequate Measures of the Effectiveness of Its Enforcement 
Strategy: 

According to OPS officials, the agency uses three types of performance 
measures to determine the effectiveness of both its enforcement 
activities and other oversight efforts: (1) the achievement of agency 
performance goals, (2) agency inspection and enforcement activity, and 
(3) the integrity management performance of pipeline operators, such as 
pipeline repairs made in response to the agency's new 
requirements.[Footnote 19] (See table 2.) These measures provide useful 
information about the agency's efforts to improve pipeline safety. For 
example, measures of pipeline repairs made in response to the agency's 
integrity management requirements provide information on the 
intermediate outcomes, or the direct results, of this new regulatory 
approach and help demonstrate how this approach leads to reductions in 
pipeline accidents and spills. However, OPS's current measures do not 
clearly indicate the effectiveness of its enforcement strategy because 
they do not measure the intermediate outcomes of enforcement actions 
that can contribute to pipeline safety, such as improved compliance, 
fewer repeat violations of specific standards, or the implementation of 
safety improvements required to correct identified problems.

Table 2: Enforcement Program Performance Measures that OPS Currently 
Uses and Is Considering Developing: 

Measures OPS currently uses: Type of measure: Achievement of agency 
performance goals; 
Description: Annual numbers of natural gas and hazardous liquid 
pipeline accidents and tons of hazardous liquid materials spilled per 
million ton-miles shipped.

Measures OPS currently uses: Type of measure: Inspection and 
enforcement activity; 
Description: Number of inspections completed; hours per inspection; 
accident investigations; enforcement actions taken, by type; and 
average proposed civil penalty amounts.

Measures OPS currently uses: Type of measure: Integrity management 
performance; 
Description: Annual numbers of accidents in areas covered by integrity 
management standards and of certain actions by pipeline operators in 
response to these standards, such as pipeline repairs completed and 
miles of pipeline assessed[A].

Measures OPS is considering developing: Type of measure: Management of 
enforcement actions; 
Description: The time taken to issue final enforcement actions, the 
extent to which penalty amounts are reduced, and the extent to which 
operators commit repeat violations.

Measures OPS is considering developing: Type of measure: Safety 
improvements ordered by OPS; 
Description: Actions by pipeline operators in response to corrective 
action orders, including miles of pipeline assessed, defects 
discovered, repairs made, and selected costs incurred.

Measures OPS is considering developing: Type of measure: Results of 
integrity management and operator qualification inspections; 
Description: The percentage of pipeline operators that did not meet 
certain requirements and the reduction in the number of operators with 
a particular deficiency. 

Source: GAO analysis of OPS information.

[A] OPS started collecting some of these data in 2002 but does not 
anticipate obtaining all of the information annually until 2005.

[End of table]

As part of efforts to improve its information systems, OPS is 
considering developing the following additional types of measures of 
the effectiveness of its enforcement and other oversight activities 
(see table 2): 

* Measures related to the management of enforcement actions. OPS is 
developing these new measures as part of efforts to integrate and 
modernize its inspection and enforcement databases and improve its 
handling of enforcement cases.

* Measures of safety improvements that were ordered by OPS. The agency 
has recently started to collect new data on actions by pipeline 
operators in response to corrective action orders and may also collect 
such data for safety orders in 2005, when it plans to start using these 
types of orders.

* The results of OPS's inspections of operator integrity management and 
operator qualification programs. OPS has developed new databases that 
track the safety issues identified in integrity management and operator 
qualification inspections, as well as enforcement actions. In 
subsequent inspections, inspectors will follow up on these issues and 
record their status.

Some of the measures under consideration could provide more information 
on the intermediate outcomes of the agency's enforcement strategy, such 
as the extent of repeat violations and repairs made in response to 
corrective action orders, as well as other aspects of program 
performance, such as the timeliness of enforcement actions. In 
addition, measures of the results of integrity management and operator 
qualification inspections could provide further information on the 
intermediate outcomes of these new regulatory approaches.

We have found that agencies that are successful in measuring 
performance strive to establish measures that demonstrate results, 
address important aspects of program performance, and provide useful 
information for decision making.[Footnote 20] While OPS's efforts to 
develop new measures have the potential to eventually produce better 
information on the performance of its enforcement program than is 
currently available, the agency has not fully adopted key practices for 
achieving these characteristics of successful performance measurement 
systems. The following sections discuss these characteristics and the 
extent to which OPS has fulfilled them in developing measures of 
enforcement performance.

Measures Should Demonstrate Results: 

Measures should be tied to program goals and demonstrate the degree to 
which the desired program results are achieved. These program goals 
should in turn be linked to overall agency goals. The new measures that 
OPS is considering are not based on such linkages, because the agency 
has not established goals for its enforcement program. Leading 
organizations seek to establish clear hierarchies of performance goals 
and measures that link the goals and measures for each organizational 
level to each successive level. Without such clear hierarchies, an 
agency will lack a straightforward road map showing how daily 
activities contribute to attaining agencywide goals. Although OPS is 
considering some new measures that could provide more information on 
the intermediate outcomes of its enforcement strategy, without first 
setting clear goals that identify the various direct results the agency 
seeks to achieve through enforcement, it may not choose the most 
appropriate measures of results and may not follow through in 
developing such measures. For example, although OPS adopted a more 
aggressive enforcement strategy starting in 2000, without appropriate 
goals and measures the agency cannot determine the effects of this new 
strategy on operators' compliance with its safety standards. OPS 
officials acknowledge that it is important to develop such intermediate 
goals and related measures but emphasize that it is challenging to do 
so because of the diversity of pipeline operations and the complexity 
of OPS's regulations.[Footnote 21]

Measures of program results can help hold agencies accountable for the 
performance of their programs. Congress needs information on program 
results to support its oversight of agencies and their budgets. 
Stakeholders can use this information to accurately judge program 
effectiveness. We asked a variety of pipeline safety stakeholders--
including representatives of industry; federal, state, and local 
agencies; and advocacy groups--for their views on the effectiveness of 
OPS's enforcement efforts. While many of them stated that they believe 
OPS's enforcement program has improved in recent years, many also 
stated that they could not comment on the impact of the agency's 
enforcement actions on pipeline safety. Some noted that this 
effectiveness is difficult to judge because of a lack of data.

Measures Should Address Important Aspects of Program Performance: 

For each goal, programs should select a few measures that cover key 
performance dimensions and take different priorities into account. 
While the new measures that OPS is considering cover a wider range of 
performance aspects than do its current measures, the agency may not be 
able to make sound decisions about which measures are the most 
important without first setting goals for its enforcement program. An 
agency official told us that a key factor in choosing final measures 
would be the availability of supporting data. However, the most 
essential measures, such as measures showing the enforcement program's 
progress in achieving compliance, may require the development of new 
data.

Developing appropriate performance measures requires carefully 
coordinated planning, including a systematic approach for identifying 
and refining potential measures that address various important aspects 
of program performance.[Footnote 22] OPS has not comprehensively 
examined its needs for measuring enforcement results as well as the 
results of other oversight efforts to ensure that its choice of 
measures will take into account and balance its various priorities. For 
example, OPS has developed databases that will track the status of 
safety issues identified in integrity management and operator 
qualification inspections, but has not yet developed the capability to 
centrally track the status of safety issues identified in determining 
compliance with its minimum safety standards. The results of follow-up 
by inspectors on the status of these issues are maintained at the 
regional office level but are not recorded in the agency's inspection 
or enforcement databases. Agency officials have told us that they are 
considering how to add this capability as part of an effort to 
modernize and integrate these databases and that the integrity 
management and operator qualification databases will serve as a model 
for this effort. However, the agency has not yet put in place a 
systematic integrated approach for designing measures of oversight 
performance, including enforcement performance.

Measures Should Provide Useful Information for Decision Making: 

Performance measures should provide agency managers with timely, 
action-oriented information in a format that helps them make decisions 
that improve program performance, including decisions to adjust 
policies and priorities.[Footnote 23] OPS uses its current measures of 
enforcement performance in a number of ways to oversee pipeline safety, 
including monitoring pipeline operators' safety performance and 
planning inspections. While these uses are important, they are only 
indirectly related to the management of enforcement results. Agency 
officials have made progress in this area by identifying possible new 
measures of enforcement results and other aspects of program 
performance, such as measures of the timeliness of enforcement actions 
that may prove more useful for managing the enforcement program.

Not having adequate measures limits OPS's ability to make informed 
decisions about its enforcement strategy. Although OPS has made major 
changes in its enforcement strategy in the last several years, it has 
decided on these changes with little information on the effectiveness 
of its prior strategy. Agency officials explained that they decided to 
increase the use of civil penalties and corrective action orders to 
improve public confidence in the agency's ability to enforce its 
standards, following the major pipeline incidents in Bellingham, 
Washington, and Carlsbad, New Mexico, in 1999 and 2000, respectively. 
They also noted that their decisions about enforcement policy are part 
of their overall approach for overseeing and improving pipeline safety 
and are not based on trends in performance measures.

OPS Has Increased Its Use of Civil Penalties; OPS and FAA Need Stronger 
Management Controls Over Collections: 

In response to criticism that its enforcement activities were weak and 
ineffective, OPS increased both the number and the size of the civil 
monetary penalties it assessed beginning in 2000. Pipeline safety 
stakeholders we spoke with expressed differing views on whether OPS's 
civil penalties are effective in deterring noncompliance with pipeline 
safety regulations. Most of the penalties that OPS assessed have been 
paid; however, OPS and FAA lack important management controls to ensure 
that penalties are collected.

The civil penalty results we present are mostly for OPS's enforcement 
of its minimum safety standards because OPS did not begin to enforce 
its integrity management standards until 2002.

OPS Has Recently Proposed and Assessed More and Larger Civil Penalties: 

OPS proposed and assessed more civil penalties during the past 4 years-
-under its current "tough but fair" enforcement approach--than it did 
in the previous 5 years, when it took a more lenient "partnering" 
enforcement approach. (See table 3. Also, see the previous section and 
app. II for a discussion of changes in OPS's enforcement approaches.) 
From 2000 through 2003, OPS proposed 127 civil penalties (about 32 per 
year on average) compared with 94 civil penalties (about 19 per year on 
average) from 1995 through 1999. Furthermore, of these proposed civil 
penalties, 88 were assessed from 2000 through 2003 (22 per year on 
average), whereas 70 were assessed from 1995 through 1999 (about 14 per 
year on average). During the first 5 months of 2004, OPS proposed 38 
civil penalties. While the recent increase in the number and the size 
of OPS's civil penalties occurred under the agency's new "tough but 
fair" enforcement approach, other factors, such as more severe 
violations, may be contributing factors as well.

Table 3: OPS's Proposed and Assessed Civil Penalties, 1994 through 
2003: 

Dollars in thousands.

Year: 1994; 
Number of proposed civil penalties: 69; 
Total proposed civil penalty amounts: $1,145; 
Average proposed penalty amount: $17; 
Number of assessed civil penalties: 58; 
Total assessed civil penalty amounts: $$735; 
Average assessed penalty amount: $13.

Year: 1995; 
Number of proposed civil penalties: 33; 
Total proposed civil penalty amounts: $734; 
Average proposed penalty amount: $22; 
Number of assessed civil penalties: 28; 
Total assessed civil penalty amounts: $531; 
Average assessed penalty amount: $19.

Year: 1996; 
Number of proposed civil penalties: 34; 
Total proposed civil penalty amounts: $532; 
Average proposed penalty amount: $16; 
Number of assessed civil penalties: 21; 
Total assessed civil penalty amounts: $227; 
Average assessed penalty amount: $11.

Year: 1997; 
Number of proposed civil penalties: 13; 
Total proposed civil penalty amounts: $355; 
Average proposed penalty amount: $27; 
Number of assessed civil penalties: 10; 
Total assessed civil penalty amounts: $321; 
Average assessed penalty amount: $32.

Year: 1998; 
Number of proposed civil penalties: 8; 
Total proposed civil penalty amounts: $94; 
Average proposed penalty amount: $12; 
Number of assessed civil penalties: 6; 
Total assessed civil penalty amounts: $78; 
Average assessed penalty amount: $13.

Year: 1999; 
Number of proposed civil penalties: 6; 
Total proposed civil penalty amounts: $110; 
Average proposed penalty amount: $18; 
Number of assessed civil penalties: 5; 
Total assessed civil penalty amounts: $100; 
Average assessed penalty amount: $20.

Year: 2000; 
Number of proposed civil penalties: 21; 
Total proposed civil penalty amounts: $1,294[A]; 
Average proposed penalty amount: $62; 
Number of assessed civil penalties: 17; 
Total assessed civil penalty amounts: $481[A]; 
Average assessed penalty amount: $28.

Year: 2001; 
Number of proposed civil penalties: 27; 
Total proposed civil penalty amounts: $1,690[B]; 
Average proposed penalty amount: $63; 
Number of assessed civil penalties: 22; 
Total assessed civil penalty amounts: $1,133; 
Average assessed penalty amount: $52.

Year: 2002; 
Number of proposed civil penalties: 47; 
Total proposed civil penalty amounts: $1,729; 
Average proposed penalty amount: $37; 
Number of assessed civil penalties: 30; 
Total assessed civil penalty amounts: $585; 
Average assessed penalty amount: $20.

Year: 2003; 
Number of proposed civil penalties: 32; 
Total proposed civil penalty amounts: $1,010; 
Average proposed penalty amount: $32; 
Number of assessed civil penalties: 19; 
Total assessed civil penalty amounts: $359; 
Average assessed penalty amount: $19.

Total[C]; 
Number of proposed civil penalties: 290; 
Total proposed civil penalty amounts: $8,692; 
Average proposed penalty amount: $30; 
Number of assessed civil penalties: 216; 
Total assessed civil penalty amounts: $$4,549; 
Average assessed penalty amount: $21. 

Source: GAO analysis of OPS and FAA data.

Note: Proposed penalty amounts ranged from $500 to $674,000 and 
assessed penalty amounts ranged from $500 to $400,000.

The data in this table may not be comparable to data that OPS reports. 
See the following text for a discussion.

[A] Excludes the penalty of $3.05 million that OPS proposed to Olympic 
Pipeline Company and the penalty of $250,000 OPS assessed to Shell 
Pipeline Company for the Bellingham accident. Including this 
extraordinarily large and unusual proposed penalty would skew the 
overall results by making the average penalty larger than it actually 
is. As noted above, no other proposed penalty amount exceeded $674,000.

[B] Excludes the penalty of $2.5 million that OPS proposed to El Paso 
Gas Company for the Carlsbad accident. We excluded this proposed 
penalty for the same reason that we excluded the proposed penalty for 
the Bellingham accident discussed in the preceding note.

[C] Totals may not add because of rounding.

[End of table]

Overall, OPS does not use civil penalties extensively. Civil penalties 
represent about 14 percent (216 out of 1,530) of all enforcement 
actions taken over the past 10 years. OPS makes more extensive use of 
other types of enforcement actions that require that operators act to 
correct safety violations. In contrast, civil penalties do not require 
a safety improvement, but represent a monetary sanction for violating 
safety regulations. Finally, OPS expects to make greater use of civil 
penalties for violations identified during integrity management 
inspections as it gains more experience with implementing this safety 
approach.

The sizes of the civil penalties have increased. From 1995 through 
1999, the average proposed civil penalty was about $19,000. From 2000 
through 2003, the average proposed civil penalty increased by over 132 
percent to about $45,000. Similarly, although to a lesser degree, 
assessed penalties increased. From 1995 through 1999, the average 
assessed civil penalty was about $18,000. From 2000 through 2003, the 
average assessed civil penalty increased by 62 percent to about 
$29,000.[Footnote 24] (All amounts are in current year dollars. 
Inflation was low during this period. If the effects of inflation are 
considered, the average assessed penalty for 1995 through 1999 would be 
$21,000, and the average assessed penalty for 2000 through 2003 would 
be $30,000, in 2003 dollars.)

We excluded two proposed penalties totaling over $5 million resulting 
from the Bellingham and Carlsbad incidents from our analysis because 
both were extraordinarily large (no other proposed penalty exceeded 
$674,000), and OPS, as of mid-July, had not assessed a penalty for the 
Carlsbad incident. (RSPA referred the penalty to the Department of 
Justice for judicial action.) Including these proposed penalties would 
have skewed our results by making the average penalty appear larger 
than it actually is.[Footnote 25]

For the 216 penalties that were assessed from 1994 through 2003, OPS 
assessed the penalty that it proposed 69 percent of the time (150 civil 
penalties). (See table 4.) For the remaining 66 penalties, OPS reduced 
the assessments by about 37 percent--from a total of about $2.8 million 
to about $1.7 million. However, the dollar difference between the 
proposed and the assessed penalties would be over three times as large 
had our analysis included the extraordinarily large penalty for the 
Bellingham, Washington, incident. For this case, OPS proposed a $3.05 
million penalty and had assessed $250,000 as of July 2004.[Footnote 26] 
If we include this penalty in our analysis, then over this period OPS 
reduced total proposed penalties by about two-thirds, from a total of 
about $5.8 million to about $2 million.

Table 4: Reductions in OPS's Proposed Civil Penalties, 1994 through 
2003: 

Year: 1994; 
Number of assessed civil penalties: 58; 
Number (percentage) of assessed civil penalties equal to proposed 
penalties: 34 (59%); 
Number (percentage) of assessed civil penalties less than proposed 
penalties: 24 (41%); 
Dollar (percentage) reduction from proposed penalties: $317,000 (49%).

Year: 1995; 
Number of assessed civil penalties: 28; 
Number (percentage) of assessed civil penalties equal to proposed 
penalties: 16 (57%); 
Number (percentage) of assessed civil penalties less than proposed 
penalties: 12 (43%); 
Dollar (percentage) reduction from proposed penalties: $77,100 (45%).

Year: 1996; 
Number of assessed civil penalties: 21; 
Number (percentage) of assessed civil penalties equal to proposed 
penalties: 17 (81%); 
Number (percentage) of assessed civil penalties less than proposed 
penalties: 4 (19%); 
Dollar (percentage) reduction from proposed penalties: $84,500 (48%).

Year: 1997; 
Number of assessed civil penalties: 10; 
Number (percentage) of assessed civil penalties equal to proposed 
penalties: 8 (80%); 
Number (percentage) of assessed civil penalties less than proposed 
penalties: 2 (20%); 
Dollar (percentage) reduction from proposed penalties: $6,000 (29%).

Year: 1998; 
Number of assessed civil penalties: 6; 
Number (percentage) of assessed civil penalties equal to proposed 
penalties: 6 (100%); 
Number (percentage) of assessed civil penalties less than proposed 
penalties: 0 (0%); 
Dollar (percentage) reduction from proposed penalties: $0 (0%).

Year: 1999; 
Number of assessed civil penalties: 5; 
Number (percentage) of assessed civil penalties equal to proposed 
penalties: 5 (100%); 
Number (percentage) of assessed civil penalties less than proposed 
penalties: 0 (0%); 
Dollar (percentage) reduction from proposed penalties: $0 (0%).

Year: 2000; 
Number of assessed civil penalties: 17; 
Number (percentage) of assessed civil penalties equal to proposed 
penalties: 12 (71%); 
Number (percentage) of assessed civil penalties less than proposed 
penalties: 5 (29%); 
Dollar (percentage) reduction from proposed penalties: $79,500 (18%).

Year: 2001; 
Number of assessed civil penalties: 22; 
Number (percentage) of assessed civil penalties equal to proposed 
penalties: 9 (41%); 
Number (percentage) of assessed civil penalties less than proposed 
penalties: 13 (59%); 
Dollar (percentage) reduction from proposed penalties: $377,400 (34%).

Year: 2002; 
Number of assessed civil penalties: 30; 
Number (percentage) of assessed civil penalties equal to proposed 
penalties: 25 (83%); 
Number (percentage) of assessed civil penalties less than proposed 
penalties: 5 (17%); 
Dollar (percentage) reduction from proposed penalties: $64,400 (34%).

Year: 2003; 
Number of assessed civil penalties: 19; 
Number (percentage) of assessed civil penalties equal to proposed 
penalties: 18 (95%); 
Number (percentage) of assessed civil penalties less than proposed 
penalties: 1 (5%); 
Dollar (percentage) reduction from proposed penalties: $5,000 (25%).

Year: Total; 
Number of assessed civil penalties: 216; 
Number (percentage) of assessed civil penalties equal to proposed 
penalties: 150 (69%); 
Number (percentage) of assessed civil penalties less than proposed 
penalties: 66 (31%); 
Dollar (percentage) reduction from proposed penalties: $1,010,900 (37%). 

Source: GAO analysis of OPS and FAA data.

Note: This table differs from table 3 in that table 3 shows all 
proposed penalties, whether or not they resulted in a final assessment. 
Table 4 portrays only the results of penalties that resulted in a final 
assessment. As discussed in the text, this table does not include the 
proposed and assessed civil penalties that OPS issued to Olympic 
Pipeline. If it did, the difference in the amount proposed and assessed 
in 2000 would be $2.8 million greater. Dollar amounts are rounded.

[End of table]

According to an OPS official, the agency reduces penalties, among other 
things, when the operator presents evidence that the inspector's 
finding is weak or wrong or when the pipeline's ownership changes 
during the period between the proposed and the assessed penalty. OPS's 
database does not provide summary information on why penalties are 
reduced. It was not practical for us to gather information on a large 
number of penalties that were reduced because to do so would have 
required reviewing each penalty record and discussing each penalty with 
headquarters and regional officials. As a result, we are not able to 
provide information on the most common reasons why penalties were 
reduced. To provide examples of reasons why penalties were reduced, we 
reviewed several of these penalties. OPS reduced one of the penalties 
we reviewed because the operator provided evidence that OPS inspectors 
had miscounted the number of pipeline valves that OPS said the operator 
had not inspected. Thus, the violation was not as severe as OPS had 
stated, and OPS reduced the proposed penalty from $177,000 to $67,000. 
OPS reduced another proposed penalty from $45,000 to $27,000 because 
the operator took immediate action to correct the violation. As 
indicated earlier in this report, good faith efforts to achieve 
compliance by operators are one factor that OPS must, by law, consider 
in imposing civil penalties. Because we reviewed only a few instances 
in which penalties were reduced, we cannot say whether these examples 
are typical.

Our results may be different from the results that OPS reports because 
of the way the data are organized. OPS reports an action in the year in 
which it occurred. For example, OPS may propose a penalty in one year 
and assess it in another year (and possibly collect it in still another 
year). The data for this action would show up in multiple years. Thus, 
OPS's data represent the activity that took place in any one year, but 
this presentation does not allow users to determine the extent to which 
the proposed penalties resulted in assessed penalties or whether the 
proposed penalty amounts were reduced, since these actions may be 
contained in OPS reports for different years. To better track the 
disposition of civil penalties, we associated assessed penalties and 
penalty amounts with the year in which they were proposed--even if the 
assessment occurred in a later year.

Stakeholders Expressed Differing Views on Whether OPS's Civil Penalties 
Deter Noncompliance: 

Although OPS has increased both the number and the size of the civil 
penalties it has imposed, the effect of this change, if any, on 
deterring noncompliance with safety regulations is not clear. The 
stakeholders we spoke with expressed differing views on whether OPS's 
civil penalties deter noncompliance.[Footnote 27] The pipeline industry 
officials we contacted said that to a certain extent OPS's civil 
penalties encourage pipeline operators to comply with pipeline safety 
regulations. One group of pipeline industry officials said that 
pipeline companies want to be on the record as being in compliance with 
pipeline safety regulations and therefore try to avoid any situation 
that would require OPS to issue an enforcement action. However, some 
industry officials said that OPS's enforcement actions are not the 
operators' primary motivation for safety. Instead, they said that the 
pipeline operators are motivated to operate safely because they need to 
avoid any type of accident, incident, or OPS enforcement action that 
impedes the flow of products through pipelines, hindering operators' 
abilities to provide good service to their customers. Pipeline industry 
officials also said that they want to operate safely and avoid pipeline 
accidents because such accidents negatively affect the public's 
perception of the company. In addition, other industry officials noted 
that OPS has other enforcement actions, such as corrective action 
orders, that give operators more incentive to operate safely because 
corrective action orders can cost companies much more money than civil 
penalties. For example, according to an OPS official, the corrective 
action order OPS imposed after the 1999 pipeline accident in 
Bellingham, Washington, cost Olympic Pipeline more than $100 million. 
This sum includes about $53 million to repair and replace needed 
infrastructure and an estimated $50 million in lost revenue. Finally, 
the three pipeline operators with whom we spoke indicated that any 
enforcement action would deter noncompliance with pipeline safety 
regulations because of the resulting negative publicity and the 
potential for costly private litigation against the operator.

Most of the interstate agents and representatives of their 
associations, insurance company officials, and the local representative 
from one state expressed views similar to those of the pipeline 
industry officials. They said that, to a certain extent, they believe 
civil penalties deter operators' noncompliance with regulations. For 
example, some of the interstate agents said that civil penalties--no 
matter what the amount--are a deterrent because the penalty puts the 
pipeline operator in the public eye. However, a few disagreed with this 
point of view. For example, representatives of the state associations 
and the local representative from one state said that OPS's civil 
penalties are too small to be a deterrent and that other OPS actions 
and the costs resulting from accidents, including private litigation, 
are better deterrents. As discussed earlier, the average civil penalty 
that OPS assessed from 2000 through 2003 was about $29,000.

Pipeline safety advocacy groups that we talked to also believed that 
the civil penalty amounts OPS imposes are too small to have any 
deterrent effect on pipeline operators. However, a representative from 
one of the groups thought that the threat of additional civil penalties 
from OPS should influence a pipeline operator to comply with pipeline 
safety regulations in the future.

According to economic literature on deterrence, pipeline operators may 
be deterred if they expect a sanction, such as a civil penalty, to 
exceed any benefits of noncompliance.[Footnote 28] Such benefits could, 
in some cases, be lower operating costs. The literature also recognizes 
that the negative consequences of noncompliance--such as those stemming 
from lawsuits, bad publicity, and the value of the products lost from 
accidents--can deter noncompliance along with regulatory agency 
oversight. Thus, for example, the expected costs of a legal settlement 
could overshadow the lower operating costs expected from noncompliance, 
and noncompliance might be deterred.

According to OPS, its policy since 1999 has been to make civil penalty 
information available to the public by publishing the final orders for 
all enforcement actions and to also publish all administrative actions 
on its Web site. We found that from 2000 through 2003 (the period OPS 
describes as its "tough but fair" enforcement era), OPS had posted 58 
percent of final orders involving assessed civil penalties on its Web 
site. An agency official explained that OPS has not posted the 
remaining penalties because of high staff turnover. To the extent that 
publicizing noncompliance information on the Web site does deter 
noncompliance, OPS's incomplete posting of assessed civil penalty 
information is not facilitating the achievement of this goal.

Operators Paid Full Amounts of Most Civil Penalties: 

For the 216 penalties that OPS assessed from 1994 through 2003, 
pipeline operators paid the full amounts 93 percent of the time (200 
instances) and reduced the amounts 1 percent of the time (2 instances). 
(See fig. 6.) Fourteen penalties (6 percent) remain unpaid, totaling 
about $837,000 (about 18 percent of the penalty amounts). In some 
instances, pipeline operators pay their penalties based on the 
proposed--rather than assessed--amount. Our results do not include an 
analysis of the number of penalties paid prior to assessment because 
FAA's and OPS's data lacked information necessary to complete the 
analysis.

Figure 6: Operators' Payment of Civil Penalties, 1994 through 2003: 

[See PDF for image] 

[End of figure] 

We followed up in one of the two instances in which the operator paid 
less than the assessed amount. In this instance, the operator requested 
that OPS reconsider the civil penalty, and OPS reduced the assessed 
penalty from $5,000 to $3,000 because the operator had a history of 
cooperation and OPS wanted to encourage future cooperation.

Neither FAA's nor OPS's data show why the 14 unpaid penalties have not 
been collected. To learn why, we spoke with both agencies about the 
status of these penalties and, based on the information provided, we 
determined that OPS closed 2 of the penalty cases without collecting 
the penalties, operators are appealing 5 penalties, OPS recently 
assessed 3 penalties, and OPS acknowledged that 4 penalties (totaling 
$45,200) should have been collected. For some penalties, the 
information that FAA and OPS provided about the collection status 
conflicted. For example, FAA reported to us that 2 penalties had been 
paid recently, which was not reflected in the information reported to 
us by OPS. Regarding the 4 penalties that should have been collected, 
OPS files indicated that final assessments had been made, but because 
FAA records did not include final orders, FAA lacked the information it 
needed to take collection action. After we brought these penalties to 
OPS's attention, OPS sent FAA the information it needed to pursue 
collection. As of June 2004, FAA had created accounts for these 
penalties and will begin sending balance due notices after the proper 
waiting period has expired.

We were not able to determine the extent to which operators' payments 
were timely (operators have 20 days to pay penalties) because we judged 
that the data elements in OPS's and FAA's databases were not reliable 
enough to do so. (See app. I.)

OPS and FAA Need Better Management Controls to Ensure That Penalties 
Are Collected: 

Even though most civil penalties are paid, their payment is more likely 
due to operators' willingness to pay than to FAA's or OPS's actions. 
FAA is not aware of the full range of civil penalties that it may 
ultimately be responsible for collecting because OPS does not routinely 
notify FAA of proposed or assessed civil penalties. We found that for 
the period from 1994 through 2003, FAA had no record of 44 of the 290 
civil penalties (totaling about $500,000 in assessments) that OPS had 
proposed.[Footnote 29] It is important for FAA to be aware of all 
proposed civil penalties because operators may choose to pay the 
proposed penalty rather than waiting for the final assessment. When FAA 
does not have a record of a civil penalty for which it receives a 
payment, it has to contact OPS for information about the penalty. In 
addition, if FAA does not know that OPS has assessed a civil penalty, 
it cannot act to collect the penalty if the operator does not pay on 
time.

Staff from OPS and RSPA's Office of Chief Counsel told us that they had 
not provided FAA with documentation of proposed or assessed penalties 
because each thought the other office was doing so. When we brought 
this apparent communication gap to OPS's attention, OPS agreed that it 
should provide FAA with civil penalty documentation so that FAA would 
be aware of the penalties that may be and are assessed. As of mid-July 
2004, OPS had not begun to provide the documentation.

Although OPS is responsible for enforcing pipeline safety, it does not 
monitor the extent to which FAA collects civil penalties for pipeline 
safety violations. OPS does not request or receive regular updates from 
FAA about the status of penalties or overall collections, although such 
reports are available. In addition, FAA does not routinely make 
available to OPS its reports on the status of civil penalties, although 
it does send them to RSPA's Office of Chief Counsel. We found that OPS 
was unaware that FAA prepares regular reports about penalties that are 
overdue or have been paid but not closed out. OPS does not evaluate the 
civil penalty data that it maintains in its enforcement database or 
review the data to ensure that its information about civil penalties is 
complete and up to date. OPS also does not compare its civil penalty 
data with FAA's data to identify missing or incomplete data. Finally, 
OPS does not evaluate its own enforcement database to identify overdue 
penalties and check with FAA on their status. After we brought these 
issues to OPS's attention, OPS officials told us that OPS is looking 
into setting up a system to monitor case activities and notify OPS and 
the deputy chief counsel when it is time to move a case to the next 
step.

Conclusions: 

OPS has been focusing much of its effort on safety initiatives in areas 
other than the enforcement of minimum standards, such as its integrity 
management program and operator qualification standards, because the 
agency believes that these initiatives will result in major 
improvements in the overall safety of the pipeline industry. In light 
of the progress OPS has made in these other areas and the issues we 
raised with OPS in preparing this report, OPS has indicated that it 
will devote more attention to managing its enforcement program than it 
has previously. However, because OPS cannot measure the effects of 
changes in its enforcement strategy on operators' performance, it will 
not know whether any management changes it makes lead to improvements 
in the industry's compliance. Without goals for its enforcement 
program, a well-defined strategy for achieving these goals, and 
performance measures linked to program goals, OPS cannot demonstrate 
how its enforcement efforts contribute to pipeline safety or learn from 
changes in its enforcement policy.

Although operators pay the vast majority of the civil penalties that 
OPS proposes or assesses, their compliance is more likely due to their 
willingness to pay than to OPS's or FAA's efforts because neither 
agency has been providing the other with the information needed to 
ensure effective penalty collections. If FAA does not know that OPS has 
imposed civil penalties, it cannot take actions to collect them, and if 
FAA does not communicate the status of its collections to OPS, OPS 
misses opportunities to understand the effects of its enforcement 
actions on operators' behavior. Finally, OPS's incomplete 
implementation of its policy to post its civil penalty actions on its 
Web site limits the public's ability to understand the enforcement 
actions that OPS has taken.

Recommendations for Executive Action: 

We are making a total of six recommendations; three to improve OPS's 
enforcement strategy and three to improve management controls over the 
collection of civil penalties.

To improve OPS's ability to determine the effectiveness of its 
enforcement strategy and make adjustments to this strategy as needed, 
we are recommending that the Secretary of Transportation direct the 
Associate Administrator for Office of Pipeline Safety to take the 
following three actions: 

* OPS should establish goals for its enforcement program.

* OPS should fully define its strategy for achieving these goals.

* OPS should establish a systematic approach for designing performance 
measures that incorporates identified key practices.

We also recommend that the Secretary of Transportation direct the 
Associate Administrator, OPS, and the Administrator, FAA, as 
appropriate, to take the following three actions to improve management 
controls over the collection of civil penalties and the public 
dissemination of information on enforcement actions: 

* OPS should inform FAA of all proposed and assessed civil penalties so 
that FAA can carry out its collection functions.

* FAA should share its reports on collections with OPS so that OPS will 
know the status of civil penalty enforcement actions.

* OPS should post all enforcement actions on its Web site, consistent 
with its policy. 

Agency Comments and Our Evaluation: 

In commenting on a draft of this report, the Associate Administrator 
for Pipeline Safety in RSPA and other officials told us that OPS 
welcomed the insights provided by the report and generally concurred 
with the report and its recommendations. The associate administrator 
emphasized that OPS enforcement has improved since 1999 through use of 
its full range of enforcement tools, including civil penalties to 
punish violators, corrective action orders to address immediate and 
potential safety concerns, and notices of amendment to require and 
monitor changes to safety programs. OPS told us that it continues to 
seek constructive interactions with the industry through the integrity 
management program; however, companies breaking the law must expect to 
be punished, and OPS will use all of its enforcement authority to 
achieve 100 percent safety compliance.

Regarding the effectiveness of OPS's enforcement strategy, OPS told us 
that, to help attain its primary performance goals (reducing the number 
of accidents by 5 percent annually and reducing the number of pipeline 
spills by 6 percent annually), it will establish intermediate outcomes 
as discussed in the report to help improve its ability to evaluate the 
effectiveness of specific enforcement tools. The officials also 
indicated that OPS expects to learn more about enforcement 
effectiveness through its two statutorily mandated technical advisory 
committees, which are reviewing, among other things, proposed 
modifications and improvements to the enforcement program. These 
committees were established pursuant to the Natural Gas Pipeline Safety 
Act of 1968 and the Hazardous Liquid Pipeline Safety Act of 1979, in 
part to serve as a sounding board for discussing pipeline safety policy 
issues. We are pleased with OPS's constructive response to our draft 
report and that the department plans to implement our recommendations 
on establishing goals for its enforcement program, defining a strategy 
for achieving those goals, and establishing a systematic approach to 
designing performance measures.

Regarding its use of civil penalties, OPS indicated that it plans to 
automate enforcement tracking to better ensure consistent application 
of policy across regional offices and improve management controls over 
the collection of civil penalties. To improve management controls over 
civil penalties collections, OPS explained that it envisions a solution 
using information technology to provide a transparent and real time 
tracking of civil penalty assessment activity between FAA, RSPA's 
Office of Chief Counsel, and OPS. We are pleased that until these 
enhancements can be deployed OPS has agreed to take the steps we 
recommended to improve management controls over the collection of civil 
penalties.

We are sending copies of this report to congressional committees and 
subcommittees with responsibility for transportation safety issues; the 
Secretary of Transportation; the Administrator, RSPA; the 
Administrator, FAA; the Associate Administrator, OPS; and the Director, 
Office of Management and Budget. We will also make copies available to 
others upon request. This report will be available at no charge on the 
GAO Web site at [Hyperlink, http://www.gao.gov].

If you have any questions about this report, please contact me at 
[Hyperlink, siggerudk@gao.gov] or James Ratzenberger at [Hyperlink, 
ratzenbergerj@gao.gov]. Alternatively, we can be reached at (202) 
512-2834. Staff who made key contributions to this report are listed in 
appendix V.

Sincerely,

Signed by:

Katherine A. Siggerud 
Director, Physical Infrastructure Issues: 

Congressional Recipients: 

The Honorable John McCain:  
Chairman: 
The Honorable Ernest F. Hollings: 
Ranking Minority Member: 
Committee on Commerce, Science and Transportation: 
United States Senate: 

The Honorable Don Young: 
Chairman: 
The Honorable James L. Oberstar: 
Ranking Democratic Member: 
Committee on Transportation and Infrastructure: 
House of Representatives: 

The Honorable Joe Barton: 
Chairman: 
The Honorable John D. Dingell: 
Ranking Minority Member: 
Committee on Energy and Commerce: 
House of Representatives: 

Appendixes: 

Appendix I: Scope and Methodology: 

To evaluate the effectiveness of the Office of Pipeline Safety's (OPS) 
enforcement strategy, we identified key elements of effective program 
management by reviewing our products on this subject, Office of 
Management and Budget guidance, and studies by the National Academy of 
Public Administration and the Urban Institute. We then determined the 
extent to which the office's strategy incorporates these three elements 
(clear program goals, a well-defined strategy for achieving goals, and 
measures of performance that are linked to program goals). For each 
element, we obtained information from OPS on its activities and plans 
and compared this information to the published criteria. We also 
reviewed the Web sites of selected regulatory agencies to determine how 
these other agencies measure enforcement results. As part of this work, 
we monitored OPS's efforts to develop a strategy that applies to all 
its enforcement activities and to improve its performance measurement 
capabilities. We supplemented these activities by interviewing pipeline 
safety stakeholders to obtain their views on the effectiveness of OPS's 
enforcement efforts. These stakeholders included industry trade 
associations, federal agencies, state agencies and associations, a 
local representative from Virginia, and pipeline safety advocacy 
groups. (Stakeholders that we contacted for this and other aspects of 
our work are listed at the end of this appendix.)

To examine OPS's civil penalty actions, we reviewed legislation; OPS 
regulations; and OPS manuals, guidelines, and protocols setting forth 
OPS's legal authority and policies and procedures for implementing this 
authority. We also obtained information from OPS, the Research and 
Special Programs Administration (RSPA), and the Federal Aviation 
Administration (FAA) about pipeline safety enforcement policies and 
procedures. (Since 1993, FAA's general accounting division has managed 
the accounts receivable for OPS's pipeline safety and RSPA's hazardous 
materials programs.) We discussed FAA's collection activities with OPS, 
RSPA's Office of Chief Counsel, and FAA officials.

In developing information about OPS's use of civil penalties, we 
analyzed civil penalty data from OPS's enforcement database and FAA's 
civil penalty receivables and collections database. Because neither 
OPS's nor FAA's civil penalty data were complete, we combined them into 
a single database. The data were incomplete in three ways. First, the 
OPS data set included penalties that were not in the FAA data set. 
Second, each data set had fields (variables) not contained in the other 
data set. Third, certain fields were common to both data sets, but the 
data for a particular enforcement action were present in one data set 
and missing in the other data set. For such variables, we substituted 
the available data for the missing data when possible. In addition, OPS 
and FAA sometimes used different methods for numbering penalties. We 
converted all penalty numbers to a standard format so we could combine 
the two data sets. When data were inconsistent with other logically 
related data, we corrected the data using our best judgment. For 
example, if OPS data showed that the assessed amount was $75,000, but 
both OPS and FAA data showed that the proposed amount was $10,500 and 
FAA data showed that both the final order and collected amount was 
$7,500 without any notation for the difference between $7,500 and 
$75,000, we assumed that the $75,000 amount was a data entry error and 
corrected the assessed amount to be consistent with the collected 
amount. When possible, we supplemented data in the two data sets with 
data from other sources. For example, if OPS's Web site included a 
final notice with the assessed penalty amount, but this information was 
not captured in the data sets, we added the information to the combined 
data set. We also discussed our preliminary results with OPS and FAA, 
and their comments led to further corrections. For example, the 
combined OPS and FAA data suggested that 16 penalties had not been 
collected. After we discussed this information with OPS and FAA, they 
provided documentation showing that many fewer penalties were 
uncollected. For example, as discussed in the body of the report, FAA 
reported to us that 2 penalties had been paid recently, which was not 
reflected in the information reported to us by OPS. In addition, we 
determined that OPS closed 2 of the penalty cases without collecting 
the penalties, operators are appealing 5 penalties, OPS recently 
assessed 3 penalties, and OPS acknowledged that 4 penalties (totaling 
$45,200) should have been collected.

In determining how OPS used its enforcement options to address 
noncompliance with pipeline safety regulations, we analyzed enforcement 
data for enforcement actions opened between 1994 and 2003 from OPS's 
enforcement database and FAA's civil penalty database. OPS's 
enforcement database contains information about instances when OPS has 
taken some type of enforcement action. OPS officials acknowledged that 
OPS's enforcement database lacked complete information on penalty 
collections and indicated that FAA tracks the collection of OPS's civil 
penalties. To assess the reliability of OPS's and FAA's data, we (1) 
performed electronic testing for obvious errors in accuracy and 
completeness and (2) interviewed officials from OPS's enforcement 
office and FAA's general accounting division who are knowledgeable 
about the data and how the data were entered. We consulted regularly 
with these officials to resolve the handling of problematic data 
entries. After these actions, and after making needed corrections, we 
determined that the data were sufficiently reliable for the types of 
analyses we wanted to pursue for this report except to determine 
whether operators paid penalties in a timely manner. In this instance, 
the discrepancies between OPS's and FAA's data and between FAA's data 
and the case files we reviewed were too great for the data to be judged 
reliable. For example, we found that the collection dates in FAA's 
database did not match the hard-copy documentation for about half of 20 
cases that we reviewed by hand.

OPS identifies each enforcement action in its enforcement database with 
a unique number. This number identifies the region, the year OPS 
initiated the action, the type of operator, and some types of 
enforcement actions. OPS's numbering system had certain characteristics 
that limited our ability to analyze the data as fully as we wanted. For 
example, the system does not identify enforcement actions where OPS 
used more than one type of enforcement, such as those that have both a 
compliance order and a notice of amendment, and it does not identify 
actions that have a compliance order. Therefore, we asked OPS to 
provide additional information detailing the enforcement and 
administrative actions taken in each instance. We used these data to 
divide the enforcement data into four sets of enforcement actions: 
civil penalties, other enforcement actions, administrative actions, and 
complex actions.

As discussed earlier in this report, our reporting of enforcement 
actions differs from OPS's. Whereas we report these actions for the 
year when OPS first responded to the related violation, OPS, in its 
annual enforcement summary report, reports the enforcement actions it 
has taken during that year, regardless of when it first responded to 
the related violations. The two reporting methods are not comparable.

As a means of better understanding OPS's civil penalty process, we 
reviewed 20 civil penalty actions initiated between 1992 through 2003. 
We chose penalties that, on their face, appeared large or small, seemed 
to have gone unpaid for a long period, may have involved repeat 
offenders, or appeared to have been reduced between the assessment and 
the collection. We reviewed the case file documentation and discussed 
the penalty with headquarters and regional enforcement officials. The 
number of penalties we reviewed was not large enough (usually about two 
to three penalty actions for each criterion we used) to draw any 
insights. Reviewing these files was time consuming, and reviewing a 
larger number of files, as well as obtaining any supporting 
documentation from the regional offices that initiated them, was not 
practical.

In determining whether OPS's civil penalties deter noncompliance, we 
interviewed pipeline safety stakeholders to obtain their views on the 
deterrent effect of OPS's civil penalties. These stakeholders included 
industry trade associations, pipeline companies, state agencies and 
associations, insurance companies, a local representative from 
Virginia, and pipeline safety advocacy groups. We supplemented the 
stakeholders' comments with information from economic literature on 
deterrence. The literature on deterrence that we reviewed included: 

* A. Mitchell Polinsky and Steven Shavell, The Economic Theory of 
Public Enforcement of Law, Journal of Economic Literature, Vol. XXXVIII 
(March 2000);

* Oren Bar-Gill and Alon Harel, "Crime Rates and Expected Sanctions: 
The Economics of Deterrence Revisited," Journal of Legal Studies, Vol. 
XXX (June 2001), pp. 485-501;

* Isaac Ehrlich, "Crime, Punishment, and the Market for Offenses," 
Journal of Economic Perspectives, Vol. 10 (Winter 1996), pp. 43-67;

* Richard A. Posner, "Economic Analysis of Law." 3RD edition (1986), 
Little, Brown and Company; and: 

* Steven D. Levitt, "Why Do Increased Arrest Rates Appear to Reduce 
Crime: Deterrence, Incapacitation, or Measurement Error?" working paper 
#5286, National Bureau of Economic Research (September 1995).

Other Aspects of Our Work: 

In determining how OPS's policies and procedures have changed over 
time, we conducted activities as described above to cover the period 
1994 through 2003.

In determining whether OPS substitutes civil penalties for corrective 
action orders, we reviewed the underlying purposes of each enforcement 
action and discussed them with OPS headquarters and regional 
enforcement officials. We reviewed a very limited number of enforcement 
cases involving civil penalties, and none of these records indicated 
that one form of penalty had been substituted for another. Because of 
the substantial effort involved, it was not practical to review a large 
number of enforcement cases.

In learning how OPS's use of civil penalties compares to FAA's for air 
carriers, we used the information described above to summarize OPS's 
civil penalty information. We compared this information to similar 
information gathered under a concurrent engagement on FAA's enforcement 
activities.[Footnote 30] We chose FAA as the comparison agency because 
it is another transportation safety agency and because information was 
readily available. Because considerable time and effort are needed to 
understand agencies' enforcement policies and practices, as well as to 
collect, ensure the quality of, and analyze data, it was not practical 
to expand the comparison to other agencies.

In determining the extent to which OPS had implemented the 
recommendations involving state activities in our May 2000 report, we 
asked OPS officials to describe the actions taken to implement the 
recommendations. We then interviewed all of the interstate agents to 
determine the extent to which they believed OPS had implemented our 
recommendations. In some instances, following interviews with 
interstate agents, we discussed with OPS the overall nature of the 
interstate agents' views. In assessing the extent to which OPS had 
implemented our recommendation on civil penalties, we discussed with 
OPS how it had responded to our recommendation. To determine whether it 
had made more use of the full range of its enforcement options, we 
examined data from OPS's pipeline incident processing enforcement 
system database and FAA's civil penalty receivables and collections 
database. We analyzed the data to determine the degree to which OPS 
used enforcement and administrative actions from 1995 through 2000 and 
compared these results with the use of these actions from 2000 through 
2003.

Finally, in comparing changes in OPS's enforcement actions with 
industry and economic trends, we interviewed OPS officials to determine 
factors that they said influenced the trends in the number of 
enforcement actions they took from 1994 through 2003. We also asked 
pipeline safety stakeholders to identify factors that might have 
influenced OPS's enforcement and administrative actions during this 
period and used our own knowledge of the area to select others. The 
factors identified are those discussed in appendix IV. We then examined 
data from OPS's pipeline incident processing enforcement system 
database and FAA's civil penalty receivables and collections database 
by analyzing the trends in the number of enforcement and administrative 
actions that OPS took from 1994 through 2003 and comparing these data 
visually with the selected factors. We obtained data from OPS on 
pipeline accidents, pipeline mileage, and OPS's inspection activities. 
We obtained data on natural gas and petroleum consumption from the 
Energy Information Administration. We obtained data on new construction 
trends from the Census Bureau.

Organizations Contacted: 

Department of Transportation: 

Federal Aviation Administration: 
Office of Inspector General: 
Research and Special Programs Administration: 
Research and Special Programs Administration: Office of Pipeline 
Safety: 
Research and Special Programs Administration: Office of the Chief 
Counsel: 

Other Federal Agencies: 

Environmental Protection Agency: 
National Transportation Safety Board: 

Pipeline Industry: 

American Gas Association: 
Interstate Natural Gas Association of America: 
American Petroleum Institute: 
Association of Oil Pipe Lines: 
Columbia Gas Transmission Corporation: 
Hawaiian Electric Company: 
Williams Midstream: 

Interstate Agents: 

Arizona Corporation Commission: 
California Department of Forestry and Fire Protection, State Fire 
Marshal: 
Connecticut Department of Public Utility Control: 
Iowa Utilities Board: 
Michigan Public Service Commission: 
Minnesota Department of Public Safety: 
New York Public Service Commission: 
Public Service Commission of West Virginia: 
Public Utilities Commission of Ohio
Virginia State Corporation Commission: 
Washington Utilities and Transportation Commission:: 

State Regulatory Associations: 

National Association of Pipeline Safety Representatives: 
National Association of Regulatory Utility Commissioners: 

Pipeline Safety Advocates: 

Citizens for Safe Pipelines: 
Cook Inlet Keeper: 
Pipeline Safety Trust: 
SAFE Bellingham: 

Insurance Companies: 

AIG Global: 
Aon Risk Services: 

Other: 

City Attorney, City of Fredericksburg, Virginia: 
Northwestern University, Evanston, Illinois:

[End of section]

Appendix II: Other Pipeline Safety Enforcement Issues: 

To address other issues required under the 2002 pipeline safety act and 
additional issues of interest to you, we examined (1) how OPS's 
enforcement policies and procedures have changed since 1990, (2) 
whether OPS substitutes corrective action orders for civil penalties, 
and (3) how OPS's policies and enforcement actions compare with those 
of FAA.

OPS's Enforcement Policies and Procedures Have Changed Since 1990: 

OPS's enforcement approach has evolved as its policies and procedures 
have changed. OPS policies have gone through three phases since 1990: 
(1) the standard inspection phase (1990 through 1994), (2) the risk 
management demonstration phase (1995 through 1999), and (3) the 
integrity management phase (2000 to the present).

* Standard inspection phase--During this phase, OPS enforced its 
minimum safety standards, conducting what it called standard 
inspections, to ensure that each pipeline operator complied with each 
pipeline safety regulation. OPS trained inspectors to complete 
inspection forms that covered all operations, but did not differentiate 
between high-risk and low-risk requirements. Individual OPS inspectors 
primarily conducted inspections on a unit basis.[Footnote 31] OPS used 
all enforcement options.

* Risk management demonstration phase--During this phase, OPS still 
focused most of its resources on enforcing minimum safety standards, 
but it also began to encourage individual operators to focus their 
resources on the greatest risks to their pipeline systems. OPS also 
began to use teams of OPS inspectors to evaluate an operator's entire 
pipeline system. The inspection goal was to determine whether the 
operator had any systemic safety issues that it needed to address. OPS 
emphasized partnering with the pipeline operators to improve pipeline 
safety rather than punishing noncompliance. As a result, OPS issued 
fewer civil penalties and more administrative actions to address 
noncompliance.

* Integrity management phase--In this phase, OPS shifted its focus from 
enforcing minimum safety standards to more comprehensive inspections of 
pipeline operators, known as the integrity management program. As a 
result, OPS conducted fewer inspections because each inspection took 
more time and covered more miles of pipeline than a standard 
inspection. However, the integrity management inspections identified 
more violations than if OPS had continued inspections of the more 
established minimum safety standards.[Footnote 32] OPS concentrated its 
enforcement actions on ensuring that operators' risk identification and 
mitigation procedures were sufficient and primarily relied on its 
notices of amendment to give operators experience in implementing the 
complex regulations. More recently, OPS has begun to propose civil 
penalties. OPS is also developing a new enforcement tool--the safety 
order--that encourages operators to take action to remedy safety-
related conditions. OPS plans to use the safety order to direct the 
operator to remedy safety-related conditions that could significantly 
change or restrict pipeline operations; however, unlike the conditions 
identified in a corrective action order, these would be conditions that 
did not pose an immediate threat to life or property. Tremendous 
pipeline failures in Bellingham, Washington, in June 1999 and in 
Carlsbad, New Mexico, in August 2000, and reports by the Department of 
Transportation's Inspector General[Footnote 33] and by us led OPS to 
abandon its partnering approach in favor of what it termed a "tough but 
fair" enforcement approach.

OPS Does Not Substitute Corrective Action Orders for Civil Penalties: 

According to OPS, the agency does not substitute corrective action 
orders for civil penalties because OPS levies corrective action orders 
and civil penalties for different reasons. OPS imposes a corrective 
action order on a pipeline operator when it finds a situation that 
presents an imminent hazard to life or property that needs to be 
addressed. OPS does not have to find that the operator has violated its 
regulations before issuing a corrective action order. For example, 
earlier this year, OPS issued a corrective action order that directed 
an operator to reduce operating pressure when OPS could not determine 
the cause of a pipeline failure. In contrast, civil penalties are not 
used to correct the underlying safety violations. Rather, OPS uses 
civil penalties as sanctions for violating federal pipeline safety 
regulations and uses other enforcement tools, such as corrective action 
orders, to correct safety violations. For example, in 2001, OPS 
assessed a $37,500 civil penalty against an operator that did not 
follow OPS's procedures during an annual test of the company's 
emergency shutdown system. OPS used the penalty as a sanction and took 
no further action. In this case, the operator already had a procedure 
in place but did not follow it. If OPS had found that the operator's 
procedures were inadequate, it could have both issued a notice of 
amendment requiring the operator to bring its procedures in line with 
OPS's regulations and imposed a civil penalty as a sanction for having 
inadequate procedures.

OPS and FAA Have Different Regulations and Processes for Using Civil 
Penalties: 

Both OPS and FAA, in its oversight of the aviation industry, use civil 
penalties, among other enforcement actions, to deter noncompliance with 
their safety regulations.[Footnote 34] While OPS regulates and may 
issue civil penalties to operators of pipeline systems, FAA regulates 
and may issue civil penalties to aircraft operators, airports, and 
individuals involved in air transport, such as pilots and 
mechanics.[Footnote 35]

Both OPS's and FAA's processes for issuing civil penalties allow for 
due process, through opportunities afforded to regulated entities to 
present evidence that may lead the regulator to reduce or withdraw the 
penalties. (See table 5.) The maximum penalties that each agency can 
impose differ significantly. OPS may impose penalties for pipeline 
operators up to a statutory maximum of $100,000 per day per violation 
up to a statutory maximum of $1 million per case, whereas FAA may 
impose much smaller penalties for aircraft operators--up to $11,000 per 
violation of federal aviation regulations or up to $30,000 per 
violation of RSPA's hazardous materials regulations. However, FAA has 
no statutory maximum penalty per case.

Table 5: OPS's and FAA's Regulations and Processes for Issuing Civil 
Penalties: 

Regulation/process: Type of safety regulations; 
OPS: Pipeline safety regulations include requirements for the 
operation, maintenance, and construction of pipelines; management of 
pipeline integrity; operator qualifications; and recordkeeping and 
reporting; 
FAA, as it applies to selected aircraft operators: Federal aviation 
regulations include requirements for aircraft maintenance, flight 
operations, qualifications of personnel, and recordkeeping and 
reporting. FAA also enforces RSPA's regulations for handling hazardous 
materials to the extent that these regulations relate to air 
transportation.

Regulation/process: Process for identifying violations; 
OPS: Inspections by OPS or state officials or accident investigations; 
FAA, as it applies to selected aircraft operators: Inspections by FAA 
officials or their designees, including air traffic controllers.

Regulation/process: Maximum civil penalty; 
OPS: $100,000 per day per violation up to a maximum of $1 million per 
case; 
FAA, as it applies to selected aircraft operators: $11,000 per 
violation of federal aviation regulations, with no maximum amount per 
case; $30,000 per violation of hazardous materials regulations, with no 
maximum amount per case.

Regulation/process: Process for proposing penalties and determining 
penalty amounts; 
OPS: Inspector identifies a potential violation and regional director 
proposes a penalty amount, taking into account factors mandated in 
statute; OPS sends a notice of probable violation, with a proposed 
civil penalty, to pipeline operator; The operator may pay the penalty 
or provide additional information and may request a hearing. Evidence 
presented by the operator may cause the penalty to be affirmed, 
reduced, or withdrawn; 
FAA, as it applies to selected aircraft operators: Inspector 
identifies a violation and recommends a civil penalty amount to 
regional program and legal offices, using FAA's enforcement sanction 
guidance. The regional program and legal offices review the proposed 
penalty and may change it; FAA notifies the air carrier of the initial 
penalty amount.[A]; The air carrier may pay the penalty or confer 
informally with FAA about the case. Evidence presented by the air 
carrier may cause the penalty to be affirmed, reduced, or withdrawn.

Regulation/process: Process for assessing and collecting penalties; 
OPS: OPS issues a final order assessing a civil penalty to the 
pipeline operator; The pipeline operator may request reconsideration of 
the penalty by RSPA's chief counsel; The pipeline operator pays the 
penalty. FAA acts as OPS's collection agent; 
FAA, as it applies to selected aircraft operators: FAA issues a final 
order or letter assessing a civil penalty to the air carrier; The air 
carrier may appeal the penalty, and a hearing will be conducted; The 
air carrier pays the penalty. FAA acts as its own collection agent. 

Source: GAO analysis of OPS and FAA information.

[A] In situations where the proposed penalties exceed $50,000 or 
involve suspending or revoking an entity's certificate, the 
headquarters legal office reviews the penalty. Proposed penalties that 
exceed $400,000 are sent to the Department of Justice for settlement.

[End of table]

Another difference between the two agencies' civil penalty processes is 
that FAA has more detailed guidance on setting penalty amounts than 
OPS. FAA's guidance lists types of violations with a corresponding 
range of civil penalty amounts for each and also lists factors that 
should be considered in setting the penalty level. OPS considers broad 
statutory factors in determining civil penalty amounts and is 
developing more detailed guidance for making these determinations as 
part of its efforts to develop an enforcement policy and detailed 
internal guidelines that reflect its current enforcement strategy.

FAA issues many more civil penalties each year than OPS. (See table 6.) 
From 1994 through 2002, FAA issued more than 10 times as many civil 
penalties to aircraft operators as OPS issued to pipeline operators. 
However, the average civil penalties that FAA assessed to the aircraft 
operators and that we included in our analyses were lower than the 
average civil penalties that OPS assessed to pipeline operators 
($14,100 versus $21,300). In addition, during this period, FAA reduced 
the civil penalties it had proposed before assessing them to a much 
greater degree than did OPS. Specifically, the total assessed penalties 
that FAA issued to aircraft operators were 59 percent lower than the 
total proposed penalties ($34.7 million versus $84.0 million),[Footnote 
36] whereas the total assessed penalties that OPS issued to pipeline 
operators were 19 percent lower than the total proposed penalties ($4.2 
million versus $5.2 million). Our comparison of OPS's and FAA's use of 
civil penalties was designed to provide some descriptive information 
but not to evaluate the two agencies' use of these penalties or to 
investigate the reasons for any differences.

Table 6: OPS's and FAA's Use of Civil Penalties, 1994 through 2002: 

Average number of civil penalties assessed per year; 
OPS: 22; 
FAA, as civil penalties apply to selected aircraft operators: 285.

Average/median civil penalty assessed per year; 
OPS: $21,300/$8,500; 
FAA, as civil penalties apply to selected aircraft operators: $14,100/
$5,000.

Percentage reduction from proposed to assessed penalties for the 9-year 
period[A, B]; 
OPS: 19%; 
FAA, as civil penalties apply to selected aircraft operators: 59%. 

Source: GAO analysis of OPS and FAA enforcement data.

Note: We did not include data for 2003 in this analysis because the FAA 
data available to us included only closed civil penalty enforcement 
actions and the number of FAA civil penalties would have been smaller 
in that year if we had included it, since some 2003 civil penalty 
actions have not yet been closed. OPS's data include open and closed 
civil penalty enforcement actions.

[A] Includes all enforcement actions with both proposed and assessed 
penalties and was calculated by determining the percentage reduction 
from the total penalties proposed for the 9-year period to the total 
penalties assessed for that period. The resulting figure for OPS--19 
percent--differs from the 37 percent figure reported for OPS in table 4 
because this latter figure indicates the percentage reduction only for 
the 31 percent of proposed civil penalties that OPS reduced. By 
contrast, the table above compares the overall reductions that OPS and 
FAA made in the penalties that they initially proposed. We did not 
compare OPS's and FAA's civil penalty collections because we lacked 
comparable data for such an analysis.

[B] FAA uses the terms "initial" and "final" rather than "proposed" and 
"assessed" when referring to civil penalties. For simplicity, this 
report refers to FAA's initial and final penalties as proposed and 
assessed penalties, respectively.

[End of table]

[End of section]

Appendix III: OPS's Implementation of Recommendations Made in Our May 
2000 Report: 

In May 2000, we made three recommendations to the Secretary of 
Transportation to improve OPS's pipeline safety program.[Footnote 37] 
Two of the recommendations proposed wider use of interstate agents and 
the third dealt with OPS's use of civil penalties. We found that OPS 
implemented two of these recommendations and implemented the intent of 
the third.

OPS Has Involved Interstate Agents More in Federal Pipeline Safety 
Activities: 

In response to our recommendation that OPS work with state pipeline 
safety officials to determine how best to involve them in federal 
pipeline safety activities, OPS told us that it had modified its 
interstate pipeline oversight program to allow more opportunities for 
state participation. OPS informed us that the 11 qualified states may 
inspect the construction of new pipelines, oversee rehabilitation 
projects and integrity management programs, investigate accidents, 
conduct inspections, and participate in nonregulatory program 
initiatives. In addition, according to OPS, states that do not qualify 
as interstate agents may apply to participate in specific, short-term 
activities such as inspecting the construction of a new pipeline or 
investigating a pipeline accident.

We contacted all 11 interstate agents to determine the extent to which 
they participate with OPS in implementing federal pipeline safety 
efforts. Ten of the 11 interstate agents told us that OPS was 
implementing our recommendation as OPS said it was doing. These 10 
states said they assisted OPS by participating in at least one of the 
activities mentioned above. The eleventh state said that OPS had not 
changed its method of involving the state; however, this state agreed 
that communication between the two parties had improved.

Although nearly all of the 11 interstate agents said that OPS was 
implementing this recommendation, 7 said OPS was too slow in letting 
them know which actions it had taken or planned to take in response to 
the potential noncompliance that the interstate agents had discovered 
during inspections. For example, two interstate agents commented that 
once they notified OPS of noncompliant activity, the case seemed to go 
into what they described as a "black hole"--indicating that they never 
heard anything else from OPS about the matter. One of these two 
interstate agents told us that it was very difficult for it to conduct 
adequate follow-up inspections (i.e., those conducted at pipeline 
companies to determine whether previously found problems have been 
corrected) without knowing what actions, if any, OPS had taken or 
planned to take. This interstate agent also noted that once, after it 
alerted OPS to noncompliant activity at one company, it found the same 
violation 2 years later during the next scheduled inspection cycle.

We brought the interstate agents' concerns to OPS's attention and, 
according to the agency, it is now providing interstate agents with 
information on the actions it took or will take in response to the 
agents' notices of noncompliant activity. OPS officials told us that 
effective November 2003, OPS began disposing of noncompliance cases in 
writing for interstate agents within 60 days after receiving notices of 
operator noncompliance from interstate agents--as required by the 
Pipeline Safety Improvement Act of 2002.

On the basis of our discussions with OPS and the interstate agents, we 
believe that OPS has implemented this recommendation.

OPS Has Involved Interstate Agents in Its Integrity Management Program: 

In response to our recommendation that OPS allow interstate agents to 
help review integrity management programs developed by the pipeline 
companies that operate in their states to ensure that these companies 
have identified and adequately addressed safety risks to their pipeline 
systems, OPS told us that it had revised its interstate agent 
agreements with qualified states to implement this recommendation.

In determining the extent to which interstate agents participate with 
OPS in reviewing integrity management plans, we contacted the six 
interstate agents that have agreements with OPS under the agency's 
hazardous liquid integrity management program.[Footnote 38] Five of 
them agreed that OPS was implementing this recommendation as OPS told 
us it was doing. The one interstate agent that did not believe OPS had 
implemented this recommendation (the same interstate agent that did not 
believe OPS had implemented our previously discussed recommendation) 
told us that while it was allowed to attend and observe one integrity 
management inspection, it was not allowed to participate--that is, it 
was not allowed to ask questions during the inspection. According to 
OPS, this was before RSPA's Chief Counsel provided interstate agents 
with verbal guidance stating that states could participate in integrity 
management inspections held outside their boundaries if the hosting 
state granted permission.

On the basis of our discussions with OPS and the interstate agents, we 
believe that OPS has implemented this recommendation.

In discussing both recommendations, we asked the interstate agents 
about the degree of partnership between them and OPS. For the first 
recommendation, 7 of the 11 interstate agents said there was or was 
close to being a true partnership with OPS. However, 3 others thought 
that better communication could improve the partnership between the two 
entities. The remaining interstate agent thought that by acknowledging 
the pipeline safety expertise that interstate agents acquired under the 
states' intrastate pipeline programs, OPS could also improve the 
partnership between the two parties. For the second recommendation, 4 
of the 6 interstate agents said there was or was close to being a true 
partnership with OPS.[Footnote 39] One of the 4 interstate agents 
thought the partnership could improve if OPS gave more advance notice 
so that interstate agents could make travel arrangements to attend 
integrity management inspections. The fifth interstate agent told us it 
did not want to offer an opinion on whether it thought a partnership 
with OPS existed. It wanted an opportunity to work with OPS on 
implementing integrity management requirements for natural gas. The 
remaining interstate agent (the same interstate agent that did not 
believe OPS had implemented either of these recommendations) thought 
there was no partnership with OPS because the agent had been allowed 
only to observe the integrity management inspections.

OPS Implemented the Intent of Our Recommendation to Examine the Effect 
of Its Reduced Use of Civil Penalties: 

We recommended that the Secretary of Transportation require that OPS 
determine whether its reduced use of civil penalties has maintained, 
improved, or decreased compliance with pipeline safety regulations. OPS 
said that it could not determine the impact of its reduced use of civil 
penalties on compliance because it did not have sufficient data to do 
so. The agency concluded that its decreased reliance on civil penalties 
did not allow it to adequately address safety concerns and was 
perceived negatively by the public and Congress. OPS subsequently 
changed its enforcement policy to make fuller use of its range of 
enforcement tools, including increasing the number and size of civil 
penalties.

While OPS did not strictly implement our recommendation, its actions to 
make fuller use of all its enforcement tools adhere to the intent of 
the recommendation. In 1994, at the very end of OPS's standard 
inspection phase, OPS issued 42 administrative actions and 95 
enforcement actions. (See fig. 7.) From 1995 through 1998, during its 
partnering phase, OPS increased its use of administrative actions, 
while issuing fewer civil penalties and non-civil-penalty enforcement 
actions. After 1998, OPS decreased its use of administrative actions. 
However, the agency did not increase its use of enforcement actions 
until 2000 when it began its "tough but fair" phase. (See app. II for 
more information on OPS's policy phases.)

Figure 7: OPS's Policy Eras and Enforcement and Administrative Actions, 
1994 through 2003: 

[See PDF for image] 

Notes: A complex action occurs when OPS takes more than one enforcement 
and/or administrative action against an operator. Generally, these 
actions include both enforcement and administrative actions. OPS 
officials told us that one of the important enforcement tools it uses 
in complex actions is the compliance order. OPS data show that the 
agency issued an average of 23 compliance orders per year from 1994 
through 2003.

This figure does not include 23 cases for which OPS data do not 
identify the associated individual actions. As a result, we were not 
able to place these cases in any of the four categories.

[End of figure] 

[End of section]

Appendix IV: OPS's Enforcement Policies Influenced Trends in 
Enforcement and Administrative Actions: 

The primary influence on trends in OPS's enforcement and administrative 
actions has been changes in OPS's enforcement policies. These policy 
changes coincided with changes in OPS's leadership. Other factors that 
contributed to OPS's policy changes and ultimately influenced trends in 
the agency's enforcement and administrative actions were two serious 
pipeline accidents and reports on them from the Department of 
Transportation's Inspector General and from us on improvements needed 
in OPS's pipeline safety program.[Footnote 40] To explore whether there 
were other possible explanations for the trends in enforcement and 
administrative actions, we analyzed trends in pipeline accidents, 
pipeline mileage, OPS's inspection activities, natural gas and 
petroleum consumption, and new construction and compared them with the 
trends in OPS's enforcement and administrative actions. We found that 
none of these data series appear to be strongly associated with the 
trends in OPS's enforcement and administrative actions.

OPS's Use of Enforcement and Administrative Actions Has Varied with 
Policy Changes: 

OPS's use of enforcement and administrative actions has evolved with 
changes in the agency's enforcement policies and leadership. As 
discussed in appendix II, OPS's enforcement policies have gone through 
three phases since 1990: (1) the standard inspection phase (1990 
through 1994), which emphasized across-the-board compliance; (2) the 
risk management demonstration phase (1995 through 1999), which focused 
on partnering with the industry to address the highest risks; and (3) 
the integrity management phase (2000 to the present), which continued 
to focus on the highest risks but also took a "tough but fair" approach 
to enforcement. OPS officials told us that the enforcement policy 
changes reflected in these three phases have been the primary influence 
on the trends in OPS's enforcement and administrative actions. In 
addition, we observed that these policy phases appear to coincide with 
changes in OPS's leadership; new associate administrators came on board 
in 1995 and mid-2000. Finally, according to OPS officials, the agency's 
latest policy change was also influenced by two major pipeline 
accidents in 1999 and 2000 that focused public and congressional 
attention on pipeline safety and led to the previously cited reports by 
the Department of Transportation's Inspector General and by us.

The changes in OPS's enforcement policies and leadership roughly 
parallel the trends in OPS's enforcement actions. (See fig. 7 in app. 
III.) As previously discussed, in 1994, at the very end of OPS's 
standard inspection phase, OPS issued 42 administrative actions and 95 
enforcement actions. From 1995 through 1998, during its partnering 
phase, OPS increased its use of administrative actions, while 
decreasing its use of civil penalties and other enforcement actions. 
After 1998, OPS's use of administrative actions decreased. In 2000, 
when OPS initiated its "tough but fair" phase, its use of enforcement 
actions started to rise.

Pipeline Accidents: 

Long-term trends in the numbers of serious pipeline accidents and in 
the accident rate for interstate hazardous liquid pipelines do not 
appear to be associated with trends in OPS's enforcement and 
administrative actions. As discussed earlier in this report, trends in 
the numbers of serious accidents for interstate natural gas and 
hazardous liquid pipelines were mixed from 1994 through 2003. (See 
figs. 2 and 3.) These trends do not parallel the wide fluctuations in 
the numbers of enforcement and administrative actions that OPS took 
during the same period. Over this same period, the accident rate for 
interstate hazardous liquid pipelines--that is, the number of serious 
accidents per billion ton-miles of hazardous liquids shipped--
decreased, while the numbers of OPS enforcement and administrative 
actions fluctuated.

For the number of all pipeline accidents per 10,000 miles of pipeline 
(where volume of products supported is not included), there appears to 
be some association between the number of accidents per 10,000 miles of 
pipeline and OPS's enforcement actions but no association between the 
number of accidents per 10,000 miles of pipeline and OPS's 
administrative actions. (See fig. 8.) This metric, like the standard 
inspections that OPS conducted from 1990 through 1994, does not take 
risk into account; it considers only the mileage of pipelines in place, 
not the amounts of products shipped--or, by implication, the risks 
involved in shipping them. The number of accidents per 10,000 miles of 
pipeline increased somewhat steadily during the period of our review, 
growing by almost 50 percent, from about 2.2 accidents per 10,000 miles 
of interstate pipeline in 1994 to 3.3 such accidents in 2002 (latest 
data available). At least through 2000, the number of accidents per 
10,000 miles of interstate pipeline and the number of OPS enforcement 
actions moved together. This parallel movement might suggest, if all 
else were equal, that OPS primarily took enforcement actions when 
serious accidents occurred. However, most OPS enforcement actions were 
the result of its routine inspections--not as a result of accident 
investigations. The number of accidents per 10,000 miles of interstate 
pipeline does not appear to coincide with the number of OPS 
administrative actions during this period.

Figure 8: OPS's Enforcement and Administrative Actions and Interstate 
Pipeline Accidents per 10,000 Miles of Pipeline, 1994 through 2003: 

[See PDF for image] 

Note: This figure does not include the complex actions included in 
figure 7 because these comprise both enforcement and administrative 
actions, and OPS's enforcement database does not provide enough 
information for us to separate the actions into these two categories. 
The figure also does not include 23 actions that OPS's database does 
not identify by type. The remaining figures in this section also 
exclude these complex and unidentified actions.

[End of figure] 

The trend in the number of serious interstate pipeline accidents (those 
causing a fatality, an injury, or $50,000 or more in property damage) 
per 10,000 miles of pipeline, like the trend in the number of all 
pipeline accidents, appears to parallel the trend in OPS's enforcement 
actions, but not the trend in OPS's administrative actions. (See fig. 
9.) During the period of our review, the number of serious interstate 
pipeline accidents rose by more than 6 percent, from 156 in 1994 to 166 
in 2003, and generally followed the same pattern as the number of 
enforcement actions. However, for serious accidents as for all 
accidents, there does not appear to be a logical connection between 
these trends. Furthermore, for serious accidents, as for all accidents, 
the trends in the number of accidents and in OPS's administrative 
actions do not appear to move together.

Figure 9: OPS's Enforcement and Administrative Actions and Serious 
Interstate Pipeline Accidents, 1994 through 2003: 

[See PDF for image] 

[End of figure] 

Pipeline Miles: 

The trend in the number of pipeline miles since 1994 does not appear to 
be associated with the trends in OPS's enforcement and administrative 
actions. (See fig. 10.) From 1994 through 2002 (latest data available), 
the miles of pipeline in the United States increased by almost 11 
percent, from almost 2.2 million miles in 1994 to more than 2.4 million 
miles in 2002. However, the numbers of enforcement and administrative 
actions that OPS issued during this period varied.

Figure 10: OPS's Enforcement and Administrative Actions and Miles of 
Pipeline, 1994 through 2003: 

[See PDF for image] 

[End of figure] 

Number of OPS Inspectors, Inspections, Inspection Days, and Types of 
Inspections: 

Our analysis shows no strong apparent association between the number of 
OPS inspectors on line, the number of inspections conducted by OPS 
inspectors, or the number of days OPS inspectors spent away from the 
office conducting inspections and the number of OPS's enforcement and 
administrative actions. (See fig. 11.) The number of OPS inspectors on 
line more than doubled from 1994 through 2003, increasing from 28 in 
1994 to 73 in 2003 at a fairly steady rate. Over the same period, the 
number of OPS's enforcement and administrative actions fluctuated 
widely.

Figure 11: OPS's Enforcement and Administrative Actions and Number of 
OPS Inspectors, 1994 through 2003: 

[See PDF for image] 

Note: This figure does not include management personnel or state 
inspectors that conduct federal inspections for OPS.

[End of figure] 

The number of inspections conducted by OPS inspectors varied from 1994 
through 2003. (See fig. 12.) The fewest inspections occurred in 1994 
and 1999 (about 730 each year), and the most took place in 1996 (almost 
1,100) and 2003 (about 1,120). These changes in the number of 
inspections do not appear to be associated with the trends in OPS's 
enforcement and administrative actions, which also varied over the same 
period but often in different directions and at different times.

Figure 12: OPS's Enforcement and Administrative Actions and Inspections 
Conducted, 1994 through 2003: 

[See PDF for image] 

[End of figure] 

The time OPS inspectors spent conducting inspections does not appear to 
be associated with the numbers of enforcement and administrative 
actions they took. (See fig. 13.) The number of days they spent away 
from the office conducting inspections more than doubled, increasing 
fairly steadily from more than 1,700 in 1994 to almost 5,300 in 2003, 
while the numbers of enforcement and administrative actions they took 
over the same period fluctuated widely.

Figure 13: OPS's Enforcement and Administrative Actions and Time Taken 
to Conduct Inspections, 1994 through 2003: 

[See PDF for image] 

[End of figure] 

Our analysis points to a possible association between the types of 
inspections conducted and OPS's enforcement and administrative 
activity. The number of standard inspections--those designed to assess 
compliance with OPS's minimum safety standards--appears to be loosely 
associated with the numbers of enforcement actions and administrative 
actions taken over the period. (See fig. 14.) The number of standard 
inspections varied from about 580 in 1994 to about 550 in 2003, peaking 
at 830 in 1996. The number of other inspections that OPS conducts, 
including construction inspections, accident investigations, and 
integrity management inspections, increased from about 150 in 1994 to 
about 560 in 2003. For many but not all of the years covered, the 
numbers of enforcement and administrative actions paralleled the 
numbers of standard and other inspections, albeit at a later date. One 
interpretation of this apparent linkage is that changes in inspection 
activity led to similar but later changes in enforcement activity. (The 
lag in enforcement activity reflects the time taken for inspectors to 
interpret their inspection results and gain management approval for any 
enforcement actions to be taken.) However, this interpretation is not 
consistent with the data for the entire 10-year period.

Figure 14: OPS's Enforcement and Administrative Actions and Inspections 
Conducted, by Type of Inspection, 1994 through 2003: 

[See PDF for image] 

[End of figure] 

Natural Gas and Petroleum Consumption: 

Trends in natural gas and petroleum consumption do not appear to be 
associated with trends in OPS's enforcement and administrative actions. 
From 1994 through 2002 (latest data available), natural gas consumption 
increased by 5.7 percent, rising from about 21.2 trillion cubic feet to 
almost 22.5 trillion cubic feet. (See fig. 15.) This trend is not 
consistent with the fluctuations in OPS's enforcement and 
administrative actions.

Figure 15: OPS's Enforcement and Administrative Actions and Natural Gas 
Consumption, 1994 through 2003: 

[See PDF for image] 

[End of figure] 

Petroleum consumption also increased, rising by almost 11 percent at a 
fairly steady rate, from almost 18 million barrels per day in 1994 to 
more than 19 million barrels per day through 2002 (latest data 
available). Over the same period, the number of enforcement and 
administrative actions that OPS issued varied widely. (See fig. 16.)

Figure 16: OPS's Enforcement and Administrative Actions and Petroleum 
Consumption, 1994 through 2003: 

[See PDF for image] 

[End of figure] 

Construction Spending: 

Historical trends in new construction do not appear to be associated 
with trends in OPS's enforcement and administrative actions. Pipeline 
accidents often result from construction activities, such as 
excavation. From 1994 through 2003, new construction steadily 
increased, as measured by Census Bureau indicators of privately owned 
housing units started and completed. For example, the number of new, 
privately owned housing units started increased by almost 27 percent at 
a fairly steady rate, from about 1.5 million in 1994 to about 1.8 
million in 2003. However, trends in OPS's enforcement and 
administrative actions varied widely over the same period.[Footnote 41] 
In particular, the trend in the number of OPS's enforcement actions 
does not appear to be associated with the trend in the number of new 
homes started. (See fig. 17.)

Figure 17: OPS's Enforcement and Administrative Actions and New 
Privately Owned Housing Units Started, 1994 through 2003: 

[See PDF for image] 

[End of figure] 

Similarly, the value of new homes completed does not appear to be 
associated with trends in OPS's enforcement activity. (See fig. 18.) 
From 1994 through 2003, in constant dollars, the annual value of 
construction put in place increased steadily by almost 23 percent, from 
about $732 billion in 1994 to about $898 billion in 2003. Over the same 
period, the number of OPS's enforcement and administrative actions 
fluctuated widely.

Figure 18: OPS's Enforcement and Administrative Actions and the Total 
Value of Construction Put in Place, 1994 through 2003: 

[See PDF for image] 

Note: Census construction value data are for public and private 
construction, published monthly, in constant (2003) dollars.

[End of figure] 

[End of section]

Appendix V: Contacts and Staff Acknowledgments: 

Contacts: 

Katherine Siggerud, (202) 512-2834 James Ratzenberger, (202) 512-2834: 

Staff Acknowledgments: 

In addition to the above, Jennifer Clayborne, Elizabeth Eisenstadt, 
Bert Japikse, Judy Guilliams-Tapia, Bonnie Pignatiello Leer, Gail 
Marnik, and Gregory Wilmoth made key contributions to this report.

(545035): 

FOOTNOTES

[1] Hazardous liquid pipelines carry products such as crude oil, diesel 
fuel, gasoline, jet fuel, anhydrous ammonia, and carbon dioxide.

[2] Standards are technical specifications that pertain to products and 
processes, such as specifications setting the size, strength, or 
technical performance required of a product, process, or material. 
National consensus standards are specifications developed by standards-
setting entities, such as the American Society for Testing and 
Materials, based on general agreement within an industry.

[3] U.S. General Accounting Office, Pipeline Safety: The Office of 
Pipeline Safety is Changing How it Oversees the Pipeline Industry, GAO/
RCED-00-128 (Washington, D.C.: May 15, 2000). This report contained 
three recommendations to make more effective use of state inspectors 
and evaluate the effect of OPS's reduced reliance on civil penalties.

[4] Before OPS imposes a civil penalty, it issues a notice of probable 
violation to the pipeline operator that documents the alleged violation 
and identifies the proposed civil penalty amount. OPS then allows the 
operator to present additional evidence. Unless the proposed violation 
and penalty are withdrawn after this step, OPS issues a final order 
that requires the operator to pay the penalty (termed "assessed 
penalties" in this report).

[5] Until February 2002, OPS required pipeline operators to report 
incidents with gross product losses of 50 barrels or more. In February 
2002, OPS reduced the reporting threshold to 5 barrels. To maintain 
consistency over the 10-year period on which we are reporting, we use 
the 50-barrel threshold for product losses after February 2002.

[6] OPS requires that operators of hazardous liquid and natural gas 
pipelines report accidents involving deaths, injuries, or property 
damage of $50,000 or more, among other things. We selected this 
indicator because these reporting requirements are common to both types 
of pipelines and because it reflects accidents with serious 
consequences.

[7] A ton-mile is 1 ton of a product shipped 1 mile. Aggregated 
industry data on the amounts of products shipped through hazardous 
liquid pipelines for 2003 are not available, so we do not present 
accident rate information for this year.

[8] For a description of challenges OPS has faced in implementing its 
integrity management approach, see U.S. General Accounting Office, 
Pipeline Safety and Security: Improved Workforce Planning and 
Communication Needed, GAO-02-785 (Washington, D.C.: Aug. 26, 2002).

[9] The operator may ask OPS to reconsider the assessed penalty amount. 
During this reconsideration, RSPA's chief counsel reviews the record 
and OPS's associate administrator renders a decision. Although OPS 
considers the evidence that operators provide to have their penalties 
reduced or withdrawn, OPS says it does not negotiate penalty amounts 
with pipeline operators.

[10] To consolidate its accounting functions, RSPA began contracting 
with FAA's general accounting division in September 1993 to collect its 
accounts receivable, including civil penalties for OPS. 

[11] In 2003, 49 state agencies, the District of Columbia, and Puerto 
Rico were certified for inspecting and enforcing regulations on 
intrastate pipelines.

[12] According to the associate administrator of OPS, recent 
interactions with the Senate Committee on Commerce, Science and 
Transportation resulting from a July 2003 pipeline rupture in Arizona 
and with us during our review have reinforced the need to devote more 
management attention to strengthening enforcement.

[13] OPS refers to the release of natural gas from a pipeline as an 
"incident" and a spill from a hazardous liquid pipeline as an 
"accident." For simplicity, this report refers to both as "accidents."

[14] We have suggested that regulatory programs develop logic models to 
develop a better understanding of how their programs deliver results in 
order to select appropriate goals. Developing such a model involves 
describing how a program's activities produce outputs and how these 
outputs are connected to intermediate and end outcomes. See U.S. 
General Accounting Office, Managing for Results: Strengthening 
Regulatory Agencies' Performance Management Practices, GAO/GGD-00-10 
(Washington, D.C.: Oct. 28, 1999). 

[15] See GAO/GGD-00-10; Agency Performance Plans: Examples of Practices 
That Can Improve Usefulness to Decisionmakers, GAO/GGD/AIMD-99-69 
(Washington, D.C.: Feb. 26, 1999); and Managing for Results: Measuring 
Program Results That Are Under Limited Federal Control, GAO/GGD-99-16 
(Washington, D.C.: Dec. 11, 1998). 

[16] We have reported on challenges that OPS faces in enforcing its 
complex integrity management program consistently and effectively. See 
GAO-02-785.

[17] U.S. General Accounting Office, Internal Control Standards: 
Internal Control Management and Evaluation Tool, GAO-01-1008G 
(Washington, D.C.: August 2001).

[18] See our previously cited reports, GAO/GGD-00-10 and GAO/GGD/AIMD-
99-69, and The Results Act: An Evaluator's Guide to Assessing Agency 
Annual Performance Plans, GAO/GGD-10.1.20 (Washington, D.C.: April 
1998).

[19] See appendix IV for trends in pipeline accidents, inspections 
conducted, and enforcement actions taken. 

[20] See, for example, GAO/GGD/AIMD-99-69 and Executive Guide: 
Effectively Implementing the Government Performance and Results Act, 
GAO/GGD-96-118 (Washington, D.C.: June 1996). We have also identified 
specific attributes of successful performance measures linked to these 
characteristics. See Tax Administration: IRS Needs to Further Refine 
Its Tax Filing Season Performance Measures, GAO-03-143 (Washington, 
D.C.: Nov. 22, 2002). 

[21] We have reported on the challenges faced by agencies in developing 
measures of program results and how they have overcome such challenges. 
See, in particular, GAO/GGD-00-10, GAO/GGD-99-16, and Managing for 
Results: Regulatory Agencies Identified Significant Barriers to 
Focusing on Results, GAO/GGD-97-83 (Washington, D.C.: June 24, 1997).

[22] See, for example, GAO/GGD-00-10; Harry Hatry, Performance 
Measurement: Getting Results (Washington, D.C.: Urban Institute, 1999); 
and National Academy of Public Administration, Designing Effective 
Performance Measures--Focus Paper (Washington, D.C.: 1999).

[23] See, for example, GAO/GGD-96-118 and U.S. General Accounting 
Office, Results-Oriented Government: GPRA Has Established a Solid 
Foundation for Achieving Greater Results, GAO-04-38 (Washington, D.C.: 
Mar. 10, 2004). 

[24] The median civil penalty was about $5,800 for 1995 through 1999 
and about $12,700 for 2000 through 2003.

[25] If these amounts are included, the average proposed penalty amount 
for the 2000 through 2003 period was about $87,600, a 350 percent 
increase over the 1995 through 1999 period. 

[26] OPS proposed a $3.05 million penalty against Equilon Pipeline 
Company, LLC (Olympic Pipeline Company) for the Bellingham incident and 
later assessed Shell Pipeline Company (formerly Equilon) $250,000, 
which it collected. According to RSPA's Office of Chief Counsel, the 
penalty against Olympic Pipeline Company is still open, waiting for 
Olympic Pipeline to emerge from bankruptcy court. 

[27] We spoke with representatives of four pipeline industry 
associations and three pipeline operators. We also spoke with all 11 
interstate agents, two state associations of utility commissioners and 
pipeline safety representatives, one local agency representative of a 
city in Virginia, and two insurance companies. Finally, we spoke with 
representatives from four pipeline safety advocacy groups. See appendix 
I for a list of those we contacted.

[28] See, for example, A. Mitchell Polinsky and Steven Shavell, The 
Economic Theory of Public Enforcement of Law, Journal of Economic 
Literature, Vol. XXXVIII (March 2000). Expected sanctions are the 
product of the sanction amount and the likelihood of being detected and 
sanctioned by that amount.

[29] We found these 44 penalties by comparing the information on 
proposed civil penalties from OPS's and FAA's databases. We identified 
44 proposed penalties in OPS's data that did not appear in FAA's.

[30] U.S. General Accounting Office, Aviation Safety: Better Management 
Controls Are Needed to Improve FAA's Safety Enforcement and Compliance 
Efforts, GAO-04-646 (Washington, D.C.: July 6, 2004).

[31] A unit is a portion of a pipeline system that can reasonably be 
evaluated in 2 to 3 days using the standard inspection process. OPS 
identifies individual inspection units on each company's pipeline 
system. The number of units in a pipeline system depends on the size of 
the system.

[32] According to OPS, there is a better chance of discovering 
noncompliant operators with the relatively new risk-based integrity 
management regulations than there is with the current minimum safety 
regulations because under the risk-based approach, safety inspectors 
look at the pipeline comprehensively rather than in segments. In 
addition, because operators are more familiar with the minimum safety 
standards, they are better able to meet them.

[33] Office of Inspector General, U.S. Department of Transportation. 
Pipeline Safety Program: Research and Special Programs Administration, 
RT-2000-069 (Washington, D.C.: Mar. 13, 2000).

[34] See GAO-04-646 for information on FAA's enforcement program.

[35] For purposes of comparing OPS's and FAA's use of civil penalties, 
we chose to limit our analysis of FAA's use of these penalties to its 
use as applied to selected aircraft operators because, as providers of 
a means of transportation, aircraft operators are more similar to 
pipeline operators than are other entities that FAA regulates. The 
aircraft operators we selected include domestic air carriers or 
commercial operators engaged in the common carriage of passengers or 
freight, as well as aircraft operators engaged in other operations, 
such as some on-demand (air taxi) operators, using U.S.-registered 
civil airplanes capable of seating 20 or more passengers or carrying a 
maximum payload of at least 6,000 pounds. We believe these operators 
most closely parallel in the aviation industry the kind of operators 
found in the pipeline industry. 

[36] See GAO-04-646 for a discussion of FAA's reduction of penalties 
from proposed amounts.

[37] GAO/RCED-00-128.

[38] We did not contact interstate agents under OPS's natural gas 
program because the natural gas integrity management rule was issued in 
December 2003--too late for operators to gain experience before we 
completed our fieldwork. 

[39] One of these 4 interstate agents thought there was a true 
partnership with OPS in implementing the integrity management program 
for small hazardous liquid pipeline operators (those with less than 500 
miles of pipeline), but not for large hazardous liquid operators 
because these inspections could have as many as three different states 
involved, and it was difficult to reach consensus. 

[40] Office of Inspector General. U.S. Department of Transportation. 
Pipeline Safety Program: Research and Special Programs Administration, 
RT-2000-069 (Washington, D.C.: Mar. 13, 2000) and GAO/RCED-00-128.

[41] We wanted to compare trends in OPS's enforcement and 
administrative actions with historical trends in population growth for 
both suburbs and inner cities. However, Census population data at this 
level of geographic detail were not available for the period covered by 
our review. 

GAO's Mission: 

The Government Accountability Office, the investigative arm of 
Congress, exists to support Congress in meeting its constitutional 
responsibilities and to help improve the performance and accountability 
of the federal government for the American people. GAO examines the use 
of public funds; evaluates federal programs and policies; and provides 
analyses, recommendations, and other assistance to help Congress make 
informed oversight, policy, and funding decisions. GAO's commitment to 
good government is reflected in its core values of accountability, 
integrity, and reliability.

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through the Internet. GAO's Web site ( www.gao.gov ) contains 
abstracts and full-text files of current reports and testimony and an 
expanding archive of older products. The Web site features a search 
engine to help you locate documents using key words and phrases. You 
can print these documents in their entirety, including charts and other 
graphics.

Each day, GAO issues a list of newly released reports, testimony, and 
correspondence. GAO posts this list, known as "Today's Reports," on its 
Web site daily. The list contains links to the full-text document 
files. To have GAO e-mail this list to you every afternoon, go to 
www.gao.gov and select "Subscribe to e-mail alerts" under the "Order 
GAO Products" heading.

Order by Mail or Phone: 

The first copy of each printed report is free. Additional copies are $2 
each. A check or money order should be made out to the Superintendent 
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or 
more copies mailed to a single address are discounted 25 percent. 
Orders should be sent to: 

U.S. Government Accountability Office
441 G Street NW, 
Room LM
Washington, D.C. 20548: 

To order by Phone: 

Voice: (202) 512-6000: 

TDD: (202) 512-2537: 

Fax: (202) 512-6061: 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 

Web site: www.gao.gov/fraudnet/fraudnet.htm
E-mail: fraudnet@gao.gov

Automated answering system: (800) 424-5454 or (202) 512-7470: 

Public Affairs: 

Jeff Nelligan, managing director,
NelliganJ@gao.gov
(202) 512-4800

U.S. Government Accountability Office,
441 G Street NW, Room 7149
Washington, D.C. 20548: