GAO-08-864R, Chemical Safety Board: Improvements in Management and Oversight Are Needed


This is the accessible text file for GAO report number GAO-08-864R 
entitled 'Chemical Safety Board: Improvements in Management and 
Oversight Are Needed' which was released on August 22, 2008.

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. 

GAO-08-864R: 

United States Government Accountability Office: 
Washington, DC 20548: 

August 22, 2008: 

Congressional Committees: 

Subject: Chemical Safety Board: Improvements in Management and 
Oversight Are Needed: 

The principal role of the Chemical Safety and Hazard Investigation 
Board (CSB) is to investigate accidental releases of regulated or 
extremely hazardous substances to determine the conditions and 
circumstances that led to the accident and to identify the cause or 
causes so that similar accidents might be prevented.[Footnote 1] 
Accidental releases of these toxic and hazardous chemicals occur 
frequently and often have serious consequences. CSB reported to 
Congress that the agency received notification of approximately 900 
chemical accidents in calendar year 2007, and that 31 of these 
accidents were serious or even fatal events that warranted the 
commitment of CSB investigators. 

CSB began operating in 1998 as an independent agency created under the 
Clean Air Act Amendments of 1990. The act directs CSB to (1) 
investigate and report on the cause or probable cause of any accidental 
chemical releases from stationary sources resulting in a fatality, 
serious injury, or substantial property damages; (2) make 
recommendations to reduce the likelihood or consequences of accidental 
chemical releases and propose corrective measures; and (3) establish 
regulations for reporting accidental releases. The agency publishes 
investigative reports and issues safety studies and videos to help 
prevent future accidents. Congress modeled CSB after the National 
Transportation Safety Board (NTSB), which has a similar public safety 
mission.[Footnote 2] Like NTSB, CSB has no enforcement authority and a 
limited regulatory role. As outlined in the authorizing statute, CSB is 
to be managed by a five-member board. Currently the board has one 
vacancy. CSB received an appropriation of $9.4 million for fiscal year 
2008 and had 39 staff as of January 30, 2008. 

In 2000, Congress asked GAO to review CSB's effectiveness in carrying 
out its mission. In our report, Chemical Safety Board: Improved 
Policies and Additional Oversight Are Needed (GAO/RCED-00-192), 
[Footnote 3] we cited problems with CSB's governance, management, 
policies, and procedures. Among other things, we recommended that CSB 
obtain the services of an existing office of inspector general (IG). 
Since fiscal year 2001, three IGs--from the Federal Emergency 
Management Agency (FEMA), the Department of Homeland Security (DHS), 
and the Environmental Protection Agency (EPA)--have provided oversight 
to CSB. EPA's IG currently provides oversight. Together, over time, 
these IGs have made 32 recommendations to address problems in 
management accountability and control, human capital management, 
compliance with its statutory requirements, and other issues. 

In response to a mandate in the Joint Explanatory Statement that 
accompanied the fiscal year 2008 Consolidated Appropriations Act, 
[Footnote 4] we examined (1) how CSB has responded to GAO and IG 
recommendations regarding CSB's investigative gap, data quality 
problems, human capital problems, and accountability and management 
problems for meeting its mission requirements and (2) the merits of the 
current oversight approach using an existing office of inspector 
general and other alternative approaches to oversight. On April 17, 
2008, we briefed staff from the House and Senate Appropriation 
Subcommittees on Interior, Environment, and Related Agencies and on May 
20, 2008, we briefed the Chairman of the House subcommittee. This 
letter summarizes the main points from our presentation. See enclosure 
II for a copy of the briefing slides from that presentation. 

To perform our review, we reviewed relevant documentation and data and 
interviewed CSB and other agency officials. We conducted this 
performance audit from October 2007 to May 2008 in accordance with 
generally accepted government auditing standards. Those standards 
require that we plan and perform the audit to obtain sufficient, 
appropriate evidence to provide a reasonable basis for our findings and 
conclusions based on our audit objectives. We believe that the evidence 
obtained provides a reasonable basis for our findings and conclusions 
based on our audit objectives. We determined that the agency's accident-
screening database is sufficiently reliable for the purpose of making 
broad estimates of the total number of accidents and accidents with 
fatalities. Among the data's limitations is the lack of quality 
controls to ensure that data are accurate and complete, especially with 
respect to fatalities. For additional information on our scope and 
methodology, see enclosure I. 

Summary of Findings: 

* CSB has implemented some GAO and IG recommendations related to 
improving its operating policies and procedures since we last reported 
in July 2000. However, we found that CSB has not fully addressed 
several critical recommendations, and problems in governance, 
management, and oversight persist. Specifically, CSB has not fully 
responded to key recommendations related to investigating more 
accidents that meet statutory requirements triggering CSB's 
responsibility to investigate, improving the quality of its accident 
data, resolving human capital problems, and ensuring accountability and 
continuity of management. 

* In our view, independent oversight from an existing IG remains the 
most effective way to help CSB address its continuing problems, 
provided that the arrangement is made permanent and funding is provided 
to the IG for the function. 

We are making recommendations to the Chairman of CSB to address 
continuing problems with governance and management. We are also 
proposing matters for congressional consideration to address continuing 
problems with oversight. Regarding the six recommendations, CSB 
generally concurred with four and disagreed with two. CSB also 
disagreed with both matters for congressional consideration. CSB did 
not generally concur with our recommendation to develop a plan to 
address the investigative gap and request the necessary resources from 
Congress to meet its statutory mandate but stated that it would work 
with Congress to clarify the issue of its statutory mandate and, if 
appropriate, seek an amendment to CSB's authorizing statute. We believe 
that CSB's willingness to work with Congress to clarify the issue of 
its statutory mandate, and if appropriate, seek an amendment to CSB's 
authorizing statute is a step in the right direction. However, CSB does 
not agree that it must currently investigate all chemical accidents 
that meet statutory requirements triggering CSB's responsibility to 
investigate. In this regard, CSB said it has "not construed the 
agency's authorizing statute as requiring investigation of every 
chemical accident involving a fatality, serious injury, or substantial 
property damage, or the potential for such consequences." We continue 
to believe the current law is clear; investigations are required for 
all accidental releases that result, or have the potential to result, 
in a fatality, serious injury, or substantial property damage. CSB is 
currently investigating far fewer accidents than is required by law and 
a plan to address the investigative gap is still necessary. 

CSB also disagreed with our recommendation to publish a regulation to 
require facilities to report information on chemical accidents. We are 
encouraged that CSB plans to publish in the Federal Register a Request 
for Information (RFI) concerning a reporting regulation to obtain the 
views and opinions of CSB's stakeholders. We recognize that this step 
could provide valuable information regarding the preparation of a 
reporting regulation. However, the request for information does not in 
itself provide assurance that CSB will follow through and issue a 
regulation as required by CSB's authorizing statute. In this regard, 
CSB said that "a reporting regulation is not needed for the narrow 
purpose of notifying the CSB of major accidents warranting the 
deployment of investigators, which appears to be the sole purpose of 
CSB's authority to issue a reporting rule." We disagree with CSB's view 
that a reporting regulation is not needed. CSB is legally required to 
promulgate a regulation. Furthermore, such a regulation would allow CSB 
to obtain more accurate, complete information to meet its statutory 
mandate. 

In addition, CSB disagreed with our matter for congressional 
consideration to consider amending CSB's authorizing statute or the 
Inspector General (IG) Act of 1978 to permanently give EPA's Inspector 
General the authority to serve as the oversight body for the agency. 
CSB questioned the independence of the EPA IG, since CSB issues 
recommendations to EPA. We understand CSB's concerns; however, the IG 
Act requires inspectors general to be independent from the agencies 
they audit and investigate, and using the EPA IG does not pose a risk 
to CSB's independent evaluation of chemical accidents.[Footnote 5] 
Finally, CSB disagreed with our matter for congressional consideration 
to provide the EPA IG with appropriations and staff allocations 
specifically for the audit of CSB via a direct line in the EPA 
appropriation. CSB said GAO did not adequately consider different 
oversight options for CSB and that oversight should be tailored to the 
size of the agency. We believe that all significant federal programs 
and entities should be subject to oversight by IGs who can provide 
sound independent audits of all significant federal operations and 
activities. The EPA IG has expertise involving the chemical management 
issues that the Board is charged with investigating, has gained 
knowledge of CSB's operations and activities in providing the Board 
with oversight over the past several years, and, like other IGs, has 
the requisite independence provided by the IG Act of 1978 necessary for 
reviewing and making recommendations to address long-standing problems 
in the Board's management performance. Given the management problems 
that our audit revealed, the need for independent IG oversight at CSB 
is especially pressing. 

CSB Has Not Fully Addressed Key Recommendations, and Problems Persist: 

CSB has not fully responded to recommendations related to its 
investigative gap, data quality, human capital, and management 
problems, and we found these problems continue. 

The Investigative Gap Persists: 

CSB has not fully responded to recommendations to address its 
investigative gap--the difference between the number of accidents it 
investigates and the accidents that meet statutory criteria triggering 
CSB's responsibility to investigate--and this gap persists. Using 
fiscal year 2002 data, the DHS IG reported in fiscal year 2004 that CSB 
deployed to 4 of 294 accidents that met statutory criteria for 
investigation. Moreover, while acknowledging that CSB lacked the 
resources to investigate all 294 accidents, the IG reported that CSB 
did not have a plan for reporting to Congress on the number and type of 
accidents it was not able to investigate, nor did it have a plan for 
narrowing the investigative gap. Consequently, the IG recommended that 
CSB develop a plan to describe and address the investigative gap and 
include the information in future budget submissions to Congress and 
the Office of Management and Budget (OMB). In response, CSB prepared a 
onetime report to Congress in 2006. CSB officials told us they did not 
repeat this report because congressional staff did not request 
subsequent annual reports with this level of detail. Rather, it 
included less detailed information in subsequent budget justifications 
to Congress. 

In fiscal year 2007, we found that CSB received notifications of 920 
chemical accidents; approximately 35 of these accidents included at 
least one fatality, and CSB investigated 1 of these. By not 
investigating all accidental releases that have a fatality, serious 
injury, substantial property damage, or the potential for a fatality, 
serious injury, or substantial property damage, CSB continues to fall 
short of its statutory mandate. CSB officials said the agency lacks the 
resources to investigate more than a small percentage of the accidents 
that meet statutory criteria triggering the board's responsibility to 
investigate. Moreover, CSB has not developed a long-term plan for 
reporting to Congress on the scope and magnitude of its investigative 
gap or a detailed strategy to address it. As a result, Congress does 
not have accurate or comprehensive information about CSB's 
investigative gap or the resources it would need to close it. 

When we compared CSB and NTSB data from 2006, we found that while 
NTSB's budget is approximately 8 times CSB's budget, NTSB investigates 
250 times as many accidents. Unlike CSB, NTSB conducts limited, office- 
based investigations that rely on the work of other entities. NTSB uses 
its statutory authority to solicit other entities' work when resources, 
or other considerations, prevent it from deploying investigators to the 
accident site, a fact that may help it better leverage its resources. 
For example, NTSB uses the work of local officials, rescue response 
units, Federal Aviation Administration (FAA) personnel, and other 
persons and organizations that might have knowledge of the accident. 
While CSB has similar statutory authority to use information gathered 
by others, the agency terminated its limited, office-based review 
program that relied on other entities work in 1999. Although the 
limited review program was less resource intensive than full 
investigations, CSB officials said that they terminated the limited 
reviews because relying on the work of other agencies conflicted with 
CSB's independence. In 2006, we reported that NTSB's use of others' 
work may present some challenges, but appears to be working well. 
[Footnote 6] Moreover, CSB has memorandums of understanding with both 
Occupational Safety and Health Administration (OSHA) and EPA that state 
that CSB may use information gathered by OSHA and EPA to aid in its 
investigation of accidents. 

Data Quality Problems Continue: 

CSB has not fully responded to IG recommendations to publish a data- 
reporting regulation and improve the quality of its accident data. As a 
result, data quality problems continue. In 2004, the DHS IG recommended 
that CSB fulfill its statutory requirement to publish a regulation for 
receiving information from facilities on their chemical accidents, and 
that CSB develop a long-term strategy to improve the quality of its 
data. Since that time, CSB has not issued the regulation, and officials 
said they have no plans to do so; instead CSB relies primarily on the 
media, such as online newspapers and television, to learn about 
chemical accidents. In addition, the DHS IG reported that CSB did not 
have adequate controls over the quality of data in the accident- 
screening database it uses to report to Congress and the public on the 
number of chemical accidents the agency screens and selects for 
investigation. The IG reported that CSB needed to monitor its data for 
completeness, accuracy, timeliness, and usefulness. 

We found that CSB lacks a long-term strategy to improve quality 
controls, and the data remain somewhat inaccurate and incomplete. For 
example, when we analyzed a subset of accidents in the database 
involving fatalities and injuries, we found at least five accidents 
(about 6 percent of the cases reviewed) where fatalities were not 
correctly recorded in the database. We also found seven accidents 
(about 4 percent of the cases reviewed) where data on injuries were 
missing as a result of incomplete data entry. Moreover, CSB does not 
have procedures to ensure that data has been entered accurately. The 
lack of data-reporting regulations and these data quality problems 
limit CSB's ability to target its resources, identify trends and 
patterns in chemical accidents, and prevent future similar accidents. 

Human Capital Problems Persist: 

CSB has not fully responded to recommendations to resolve its human 
capital problems. In 2002, the FEMA IG found that CSB had a shortfall 
of investigators and had not made hiring them a priority. In addition, 
it found that CSB lacked a central human capital manager, comprehensive 
strategic human capital plan, and performance measures and criteria. 
The FEMA IG recommended that CSB make hiring investigators a top 
priority and made several recommendations to strengthen its human 
capital planning and management. 

In response, CSB consolidated human capital responsibilities under a 
full-time human resources manager, developed several agencywide goals 
to improve human capital, and hired more investigators; however, we 
found that CSB's human capital strategy was not comprehensive, lacked a 
detailed action plan for closing the investigator shortfall, did not 
include input from staff investigators, and lacked performance 
measures--actions included in the Strategic Management of Human Capital 
portion of the President's Management Agenda. 

CSB officials told us they have difficulties attracting and retaining 
investigators. We found that more employees left CSB in fiscal years 
2006 and 2007 than were hired. In fiscal years 2006 and 2007, three of 
five investigators who left were senior investigators with 5 to 7 years 
of experience. Yet CSB hired mostly interns during the same 2 fiscal 
years. CSB hired these interns through the Federal Career Intern 
program, which is designed to attract and retain employees for federal 
service. However, some officials we interviewed, including a board 
member, investigation managers, and investigators, told us intern 
investigators are encouraged to leave CSB to gain experience in private 
industry or to pursue graduate degrees. In addition, CSB officials said 
they offered retention bonuses to high-performing mid-and senior-level 
investigators before they left the agency. However, the investigators 
declined the bonuses because these individuals said they received 
significantly higher compensation from the private sector. In addition, 
CSB officials said they did not offer retention bonuses to resigning 
intern investigators because they did not think it would make a 
difference since these individuals would be earning more in their new 
jobs. We found that the agency has not paid a retention bonus to any 
employee since September 2002. Moreover, we found that in fiscal year 
2006, CSB reprogrammed compensation funds of $627,891 to other 
priorities, including producing safety videos and redesigning its Web 
site, and that in fiscal year 2007, CSB reprogrammed compensation funds 
of $407,383 to similar purchases. 

Accountability and Management Problems Continue: 

CSB has not fully responded to recommendations to delegate the 
authority to effectively manage the day-to-day administrative functions 
to a permanent chief operating officer (COO). In March 2002, the FEMA 
IG cited fractured management, a weakened chain of command, and board 
member intervention in routine administration and recommended that the 
board delegate day-to-day administrative functions to a permanent COO 
to ensure continuity of management and accountability. In response to 
the FEMA IG's recommendation, CSB hired a COO in 2002 to effectively 
manage the day-to-day operations of CSB, but the individual left in 
2004. The board subsequently eliminated the position and transferred 
the COO's responsibilities to individual program managers and the 
board. 

We found that CSB lacks a permanent senior executive to establish 
performance goals, hold program mangers accountable for meeting those 
goals, and demonstrate improvement in the agency's ability to meet its 
statutory mandate over time. Without a COO, the agency may be unable to 
ensure continuity of performance and accountability when board members 
and chairs leave the agency. 

CSB board members said that a similar executive director position might 
reduce the administrative responsibilities of the board; however, in 
comments to GAO, CSB did not support filling a COO or similar position 
at this time because a COO is not likely to provide any additional 
management skills not already represented at the agency. The 
investigation managers and investigators we interviewed generally 
expressed support for reinstating the COO position to improve the 
continuity of administrative management. 

Inspector General Oversight Is Still Warranted: 

On the basis of our review of CSB's history and current operations; we 
believe that the independent institutional audit presence of an IG 
remains the best option for ensuring that CSB is accountable to 
Congress for meeting its statutory requirements. We reconsidered the 
three options for oversight we suggested in 2000, which include (1) 
establishing an in-house audit and investigations unit, (2) contracting 
out for evaluations of its operations and programs, and (3) obtaining 
the services of an existing office of inspector general. We determined 
that the first two options are not appropriate for the board for 
several reasons. The first option--an in-house audit unit--is not 
practical because CSB's history of management problems warrants a level 
of independent oversight that may be difficult to achieve by an 
internal audit function. In addition, the limited staffing that would 
reasonably be allocated to this function at an agency of this size 
would lack the varied expertise needed to address these problems. The 
second option--contracting out for evaluations--is not appropriate 
because of the limitations of contracting in terms of both audit 
independence and the potentially limited duration of the contracting 
relationship. CSB officials told us they prefer the second option 
because they believe CSB's small size does not justify independent 
institutional IG oversight. In addition, CSB officials said that CSB is 
classified as one of 54 federal entities defined in the IG Act of 1978 
for which the act did not provide an IG, but rather required annual 
reporting of their audit and investigative activities to Congress and 
the Office of Management and Budget. However, in our view, the CSB's 
long-standing, serious, and intractable management problems make it 
unlikely that the reporting requirements for federal entities will 
ensure that the CSB has an appropriate level of oversight to address 
its management problems. 

Given these factors, in our view IG oversight remains the most 
appropriate oversight option for CSB. Nonetheless, we recognize that 
there are some shortcomings with the current EPA IG's oversight 
relationship. First, the arrangement is not permanent, a fact that may 
undermine the continuity of oversight. Second, EPA IG officials told us 
they have no plans to conduct future program evaluations of CSB because 
they are allocating their limited evaluation resources to other 
priority issues within EPA. However, we do not believe, as CSB asserts, 
that the EPA IG's assignment and work call into question CSB's intended 
independence from EPA. According to CSB, CSB independently evaluates 
and reports to Congress on EPA programs in chemical accident 
prevention, and CSB's independence from EPA was deliberate and 
carefully considered. By statute, inspectors general are independent 
from the agencies they audit and investigate so the EPA IG must 
maintain his or her independence from EPA officials and program 
employees. With such independence, the IG poses no risk to CSB's 
independent evaluations of chemical accidents. 

Conclusions: 

After 10 years of operation, CSB continues to operate in noncompliance 
with its statutory mandates. CSB stresses that it recognizes the 
importance of its investigations to identify root causes of accidental 
releases and recommend regulatory action to prevent such accidental 
releases, but it is not investigating all chemical releases that have a 
fatality, serious injury, substantial property damage, or the potential 
for a fatality, serious injury, or substantial property damage. 

Given the resource constraints on the board that limit its ability to 
investigate all chemical accidents resulting in fatalities, serious 
injuries, or substantial property damage, it is particularly important 
that CSB better leverage its existing resources by using other 
entities' work, have the best available data on which to make decisions 
on those accidents that are most important to investigate, and use all 
available human capital tools to retain staff. Even though CSB has a 
statutory requirement to issue a regulation requiring facilities to 
report their chemical releases, the board has resisted requiring such 
reporting, preferring to rely on alternative information sources, such 
as major news organizations. Requiring facilities to report certain 
information on accidental releases would provide CSB with better data 
than it currently receives from media sources. 

The difficulties that CSB has experienced are largely the result of 
inadequate management accountability for addressing long-standing 
problems and for clearly identifying and attempting to meet CSB's staff 
requirements to perform investigations of chemical accidents. While we 
recognize that CSB may not have sufficient resources to investigate 
every accident within its purview, as NTSB reports it does, we believe 
CSB is missing opportunities to investigate more accidents and possibly 
prevent fatalities, serious injuries, and substantial property damage 
in the future. For example, other federal agencies, such as EPA and 
OSHA, collect accident information that, to the extent that the 
information meets CSB's data quality standards, could provide 
additional resources to help the board meet its mission. 

Recommendations for Executive Action: 

We recommend that the chairman of the Chemical Safety Board: 

* develop a plan to address the investigative gap and request the 
necessary resources from Congress to meet CSB's statutory mandate or 
seek an amendment to its statutory mandate; 

* consider using the work of other entities, such as government 
agencies, companies, and contractors (subject to an assessment of the 
quality of their work), to a greater extent to maximize the board's 
limited resources; 

* improve the quality of its accident-screening database by better 
controlling data entry and periodically sampling accident data to 
evaluate their consistency and completeness; 

* publish a regulation requiring facilities to report all chemical 
accidents, as required by law, to better inform the agency of important 
details about accidents that it may not receive from current sources; 

* consider reinstating the position of chief operating officer, with 
delegations of responsibility for establishing performance goals, 
holding program mangers accountable for meeting those goals, and 
demonstrating improvement in the agency's ability to meet it statutory 
mandates over time; and: 

* use the Strategic Management of Human Capital portion of the 
President's Management Agenda to provide criteria for developing a 
comprehensive human capital plan, with input from investigators that 
includes specific objectives and performance measures to improve 
accountability for results and to assist the agency in its goal of 
improving its human capital and infrastructure. 

Matters for Congressional Consideration: 

Congress may wish to consider amending CSB's authorizing statute or the 
Inspector General Act of 1978 to permanently give EPA's Inspector 
General the authority to serve as the oversight body for the agency. 

As Congress prepares the appropriation of the EPA Inspector General, it 
may wish to consider providing the Inspector General with 
appropriations and staff allocations specifically for the audit 
function of CSB via a direct line in the EPA appropriation. 

Agency Comments and Our Evaluation: 

We provided a draft of our report to CSB for its review and comment. We 
received written comments from CSB's Chairman and Chief Executive 
Officer (CEO). These comments and our detailed response to them are 
presented in enclosure III. CSB also provided technical comments, which 
we have incorporated into the report as appropriate. 

Regarding the six specific recommendations we made in the report, CSB 
generally concurred with four and disagreed with two. CSB also 
disagreed with both matters for congressional consideration we 
identified in the report. CSB generally concurred that it should (1) 
consider using the work of other entities to maximize the board's 
limited resources, (2) improve the quality of its accident-screening 
database, (3) consider reinstating the position of chief operating 
officer, (4) use the Strategic Management of Human Capital portion of 
the President's Management Agenda as a guide for developing a 
comprehensive human capital plan. 

CSB did not generally concur with our recommendation to develop a plan 
to address the investigative gap and request the necessary resources 
from Congress to meet its statutory mandate. The agency reports that it 
will seek additional investigation resources and "will draft a plan for 
obtaining information on additional chemical accidents occurring in the 
United States, and clearly set forth a risk-based approach to accident 
selection and investigation." However, it does not commit to meet CSB's 
statutory mandate for investigating chemical releases, citing its view 
that "CSB has not construed the agency's authorizing statute as 
requiring investigation of every chemical accident involving a 
fatality, serious injury, or substantial property damage, or the 
potential for such consequences." CSB has not explained the basis for 
its interpretation. The 1990 Clean Air Act Amendments establishing the 
CSB states "[t]he Board shall investigate (or cause to be 
investigated)�any accidental release resulting in a fatality, serious 
injury or substantial property damage." Further, the act states that 
"in no event shall the Board forgo an investigation where an accidental 
release causes a fatality or serious injury among the general public, 
or had [sic] the potential to cause substantial property damage or a 
number of deaths or injuries among the general public." As noted in our 
briefing, this language clearly identifies which accidental releases 
CSB is required to investigate; that is, investigations are required 
for all accidental releases that result, or have the potential to 
result, in a fatality, serious injury, or substantial property damage. 
Although CSB did not agree with this interpretation, the agency stated 
that it would work with Congress to clarify the issue of its statutory 
mandate and, if appropriate, seek an amendment to CSB's authorizing 
statute. We believe that CSB's willingness to work with Congress to 
clarify its statutory mandate is a step in the right direction. 
However, we continue to believe the current law is clear; CSB should 
meet its current mandate or seek an amendment to its authorizing 
statute. 

CSB disagreed with our matter for congressional consideration to 
consider amending CSB's authorizing statute or the IG Act of 1978 to 
permanently give EPA's Inspector General the authority to serve as the 
oversight body for the agency. CSB said that while the EPA IG is one 
option for oversight, other offices of inspector general are also 
available, some of which may be more appropriate for the role. CSB also 
questioned the independence of the EPA IG, since CSB issues 
recommendations to EPA program offices. We assessed other options for 
oversight that we had considered in our 2000 report, as well as options 
presented to us by CSB. We understand CSB's concerns; however, the IG 
Act requires inspectors general to be independent from the agencies 
they audit and investigate. The EPA IG has expertise involving the 
chemical management issues that the Board is charged with 
investigating, has gained knowledge of CSB's operations and activities 
in providing the Board with oversight over the past several years, and, 
like other IGs, has the requisite independence provided by the IG Act 
of 1978 necessary for reviewing and making recommendations to address 
long-standing problems in the Board's management performance. As a 
result, using the EPA IG does not pose a risk to CSB's independent 
evaluation of chemical accidents. 

CSB also disagreed with our matter for congressional consideration to 
provide the EPA IG with appropriations and staff allocations 
specifically for the audit of CSB via a direct line in the EPA 
appropriation. CSB said GAO did not adequately consider different 
oversight options for CSB and that oversight should be tailored to the 
size of the agency. CSB also noted that it currently obtains its 
financial and information security audits for about $60,000 a year. We 
believe that all significant federal programs and entities should be 
subject to oversight by IGs who can provide sound independent audits of 
all significant federal operations and activities. Given the management 
problems that our audit revealed, the need for independent IG oversight 
at CSB is especially pressing. Further, financial and information 
security audits are not a substitute for the oversight of program 
management provided by an independent IG. 

We are sending copies of this report to the Chairman and CEO of the 
Chemical Safety and Hazard Investigation Board, appropriate 
congressional committees, and other interested parties. We will also 
make copies available to others upon request. In addition, the report 
will be available at no charge on the GAO Web site at [hyperlink, 
http://www.gao.gov]. 

If you or your staff have any questions about this report, please 
contact me at 202-512-3841 or stephensonj@gao.gov. Contact points for 
our Offices of Congressional Relations and Public Affairs may be found 
on the last page of this report. Major contributors to this report were 
Ed Kratzer, Assistant Director; Vanessa Dillard; Brian M. Friedman; 
Angela Miles; Alison O'Neill; Michael Sagalow; Rebecca Shea; John C. 
Smith; and Jeanette Soares. 

Signed by: 

John B. Stephenson: 
Director, Natural Resources and Environment: 

Enclosures: 

Congressional Addressees: 

The Honorable Dianne Feinstein: 
Chairman: 
The Honorable Wayne Allard: 
Ranking Member: 
Subcommittee on Interior, Environment, and Related Agencies: 
Committee on Appropriations: 
United States Senate: 

The Honorable Norman D. Dicks: 
Chairman: 
The Honorable Todd Tiahrt: 
Ranking Member: 
Subcommittee on Interior, Environment, and Related Agencies: 
Committee on Appropriations: 
House of Representatives: 

[End of section] 

Enclosure I: Scope and Methodology: 

To perform our review, we analyzed authorizing statutes, regulations, 
legislative history, and GAO and office of inspector general (IG) 
reports, and other literature. We also reviewed and analyzed the 
Chemical Safety and Hazard Investigation Board's (CSB) strategic plan, 
human capital report, policies and procedures, and other program 
documents, and compared them to similar documents from the National 
Transportation Safety Board (NTSB). In addition, we reviewed a subset 
of CSB's accident screening database that included two variables CSB 
uses to rank the seriousness of an accident--fatalities and injuries 
and the narratives explaining these accidents. We also consulted GAO's 
guidance on management best practices, human capital, data reliability, 
and oversight. In addition, we interviewed officials from CSB, 
officials from past and current IGs. We also consulted NTSB officials 
and GAO experts on management, data reliability, human capital, and 
oversight issues. 

We did not evaluate the quality of CSB's investigative products issued 
to date, or the quality and effectiveness of its reports, 
recommendations, and promotion of preventive actions, because these 
evaluations were outside the scope of our mandate and are not relevant 
to the findings and conclusions of this engagement. We also did not 
conduct a detailed evaluation of the oversight arrangements for the 
other 53 federal entities for comparison with CSB because the 
continuing management problems found in our audit highlight the need 
for IG oversight. 

We conducted this performance audit from October 2007 to May 2008 in 
accordance with generally accepted government auditing standards. Those 
standards require that we plan and perform the audit to obtain 
sufficient, appropriate evidence to provide a reasonable basis for our 
findings and conclusions based on our audit objectives. We believe that 
the evidence obtained provides a reasonable basis for our findings and 
conclusions based on our audit objectives. We determined that the 
agency's accident screening database is sufficiently reliable for the 
purpose of making broad estimates of the total number of accidents and 
accidents with fatalities. Among the data's limitations is the lack of 
quality controls to ensure that data are accurate and complete, 
especially with respect to fatalities. 

[End of section] 

Enclosure II: Slides from April 17, 2008, Briefing to Congressional 
Staff: 

Chemical Safety Board: Improvements in Management and Oversight Are 
Needed: 

Briefing to the Subcommittees on Interior, Environment, and Related 
Agencies: 
Committee on Appropriations: 
U.S. Senate: 
U.S. House of Representatives: 

Background: 

The Chemical Safety and Hazard Investigation Board (CSB) is an 
independent agency that was created under the Clean Air Act Amendments 
of 1990 and began operating in 1998. The act directed CSB to: 

* investigate and report on the cause or probable cause of any 
accidental chemical releases resulting in a fatality, a serious injury, 
or substantial property damage; 

* make recommendations to reduce the likelihood or consequences of 
accidental chemical releases and propose corrective measures; and; 

* establish regulations for reporting accidental releases. 

Congress modeled CSB after the National Transportation Safety Board 
(NTSB), which has a similar mission. 

Like NTSB, CSB has no enforcement authority and a limited regulatory 
role. 

As outlined in the statute, CSB is to be managed by a five-member 
board. Currently the board has one vacancy. 

In fiscal year 2009, CSB requested $10.6 million and 47 staff. 

In fiscal year 2000, CSB requested $12.5 million and 60 staff. 

Prior year appropriations and on-board staff levels include: 

* fiscal year 2008, $9.4 million and 39 staff (as of 1/30/2008); 
* fiscal year 2006, $9.2 million and 41 staff (as of 9/30/2006); 
* fiscal year 2004, $8.2 million and 40 staff (as of 9/30/2004); and; 
* fiscal year 2000, $8.0 million and 27 staff (as of 6/15/2000). 

In 2000, GAO reported problems with CSB�s governance, management, 
policies, and procedures and recommended, among other things, that CSB 
obtain the services of an existing office of inspector general (IG). 

Since fiscal year 2001, three existing IGs have provided oversight to 
CSB: Federal Emergency Management Agency (FEMA), the Department of 
Homeland Security (DHS), and the U.S. Environmental Protection Agency 
(EPA). EPA�s IG currently provides oversight. 

From fiscal year 2002 through 2007, inspectors general have made 32 
recommendations to address problems in management accountability and 
control, human capital management, compliance with its statutory 
requirements, and other issues. 

[End of section] 

Objectives: 

You asked us to examine: 

1. how CSB has responded to GAO and IG recommendations regarding CSB�s 
investigative gap, data quality problems, human capital problems, and 
accountability and management problems for meeting its mission 
requirements and; 

2. the merits of the current oversight approach�using an existing 
office of inspector general�and other alternative approaches to 
oversight. 

Scope and Methodology: 

To answer these questions, we reviewed: 

* authorizing statutes, regulations, legislative history and GAO and IG 
reports, and other literature; 

* CSB�s accident screening database for fiscal year 2006 and fiscal 
year 2007; 

* CSB�s strategic plan, human capital report, and other program 
documentation such as policies and procedures; 

* similar program documents from NTSB for comparison to CSB documents; 

* GAO guidance pertaining to management best practices, human capital, 
data reliability, and oversight; and; 

* The mission and oversight arrangements for 54 federal entities, 
including CSB, as defined by the Inspector General Act of 1978. 

We did not review or evaluate the quality or effectiveness CSB�s 
products, recommendations or preventive actions as that was outside the 
scope of our mandate and not relevant to this engagement�s findings and 
conclusions. 

We also interviewed officials from CSB and past and current IGs and 
consulted GAO experts on management, data reliability, human capital, 
and oversight issues. 

We conducted this performance audit from October 2007 to May 2008 in 
accordance with generally accepted government auditing standards. Those 
standards require that we plan and perform the audit to obtain 
sufficient, appropriate evidence to provide a reasonable basis for our 
findings and conclusions based on our audit objectives. We believe that 
the evidence obtained provides a reasonable basis for our findings and 
conclusions based on our audit objectives. We determined that the 
agency�s accident-screening database is sufficiently reliable for the 
purpose of making broad estimates of the total number of accidents and 
accidents with fatalities. Among the data�s limitations is the lack of 
quality controls to ensure that data are accurate and complete, 
especially with respect to fatalities. 

[End of section] 

Results in Brief: 

CSB has implemented some GAO and IG recommendations related to 
improving its operating policies and procedures since we last reported 
in July 2000. However,CSB has not fully addressed several critical 
recommendations, and problems in governance, management, and oversight 
persist. Specifically, CSB has not fully responded to key 
recommendations related to (1) investigating more accidents that meet 
statutory requirements triggering the CSB�s responsibility to 
investigate,(2) improving the quality of its accident data, (3) 
resolving human capital problems, and (4) ensuring accountability and 
continuity of management. 

In our view, independent oversight from an existing IG remains the most 
effective way to help CSB address its continuing problems, provided 
that the arrangement is made permanent and funding is provided to the 
IG for the function. CSB officials disagreed based on the agency�s 
small size and its existing annual reporting requirements to Congress 
and the Office of Management and Budget. 

Therefore: 
* We are making recommendations to the Chair of CSB to address 
continuing problems with governance and management. 

* We are proposing matters for congressional consideration to address 
continuing problems with oversight. 

[End of section] 

The investigative gap persists: 

CSB has not fully responded to recommendations to address its 
investigative gap�the difference between the number of accidents it 
investigates and the accidents that meet statutory criteria triggering 
CSB�s responsibility to investigate. 

Initial problem: Using fiscal year 2002 data, the DHS IG reported in 
fiscal year 2004 that of the 613 accidents CSB screened, about 294 met 
statutory criteria, and CSB deployed investigators to 4. Moreover, 
while acknowledging that CSB lacked the resources to investigate all 
294 accidents, the IG reported that CSB did not have a plan for 
reporting to Congress on the number and type of accidents it was not 
able to investigate or a plan for narrowing the investigative gap. 

IG�s recommendation: The DHS IG recommended that CSB develop a plan to 
describe and address the investigative gap and include the information 
in future budget submissions to Congress and OMB. 

How CSB responded: Two years after the IG made the recommendation, CSB 
prepared a onetime report to Congress in 2006. The report described 14 
accidents occurring in 1 year that it said it would have investigated 
if it had had the resources. However, the report did not describe a 
plan to address the investigative gap identified by DHS IG. 

* CSB has not repeated this report; rather, it included less detailed 
information in subsequent budget justifications to Congress. 

* CSB officials told us that congressional staff did not request 
subsequent annual reports with this level of detail. 

* CSB officials also said the agency lacks the resources to investigate 
more than a small percentage of the accidents that meet statutory 
criteria triggering the board�s responsibility to investigate. 

Problem remaining: CSB continues to underestimate the number of 
accidents it would need to investigate to meet statutory requirements. 
In addition, it has not significantly narrowed its investigative 
gap�investigating about 6 accidents each year from fiscal years 2004 
through 2007. Moreover, CSB has not developed a long-term plan for 
reporting to Congress on the scope and magnitude of its investigative 
gap or a detailed strategy to address it. Therefore: 

* CSB continues to fall short of its statutory mandate, and accidents 
that involve fatalities, serious injuries, and substantial property 
damage go uninvestigated. 

* Congress does not have accurate or comprehensive information about 
CSB�s investigative gap or the resources it would need to close it. 

In fiscal year 2007, CSB investigated less than 1 percent (5 of 920) of 
accidents of which it was notified. 

Examples of accidents in fiscal years 2006 and 2007 that CSB did not 
investigate include: 

* An oil well explosion that killed 3 teens. 

* A methanol flash fire at a school that injured 8 students and their 
teacher. Three students and the teacher were hospitalized for burns on 
their upper torsos, faces, and hands. 

* A natural gas well explosion that killed 1 worker and forced hundreds 
of residents out of their homes for hours. 

* A propane explosion that killed 3 workers and injured 47. 

* A waste-processing plant that released noxious chemicals and may have 
sickened more than 200 people. 

Table 1: Information on Chemical Accidents and Investigations, Fiscal 
Years 2006 and 2007: 

Total accidents CSB recorded in its screened database: 
2006: 822[A]; 
2007: 920[A]. 

Accidents CSB deployed to: 
2006: 9[B]; 
2007: 8[B]. 

Accidents CSB investigated: 
2006: 6; 
2007: 5. 

Accidents with fatalities: 
2006: 38[A]; 
2007: 35[A]. 

Accidents with fatalities that CSB investigated: 
2006: 5; 
2007: 1. 

Source: GAO analysis of CSB data. 

[A] Because of data quality limitations, this total number of accidents 
and accidents with fatalities identified and reported in the database 
may be underestimated. 

[B] The number of deployments is greater than the number of accident 
investigations because CSB deployed to accidents it did not ultimately 
investigate. 

[End of table] 

CSB and NTSB are both required to investigate all accidents meeting 
certain criteria. NTSB is required to investigate (1) all civil 
aviation accidents; (2) all pipeline accidents in which there is a 
fatality, substantial property damage, or significant injury to the 
environment; (3) all railroad accidents in which there is a fatality or 
substantial property damage, or that involves a passenger train; and 
(4) certain major marine casualties. Likewise, CSB is required to 
investigate all accidental chemical releases into the air from 
stationary sources that cause a fatality, serious injury, or 
substantial property damage. NTSB does have discretion in selecting 
which highway and "other modes of transportation" accidents to 
investigate. CSB's authorizing statute does not give it any discretion 
in selecting which accidents to investigate. 

While NTSB�s budget is approximately 8 times CSB�s budget, NTSB 
investigates 250 times as many accidents. 

Figure 1 describes the annual budget and the number of investigations 
CSB and NTSB started in fiscal year 2006. 

Figure 1. CSB and NTSB Number of Accident Investigations Initiated and 
Budget, fiscal year 2006: 

[See PDF for image] 

This figure is a vertical bar graph depicting the following data: 

Investigations: 
CSB: 6; 
NTSB: 1,521. 

Budget: 
CSB: $9.1 million; 
NTSB: $75.9 million. 

Source: GAO analysis of CSB and NTSB data. 

Note: The total of 1,521 is aviation accident investigations only. 
Nonaviation accident investigations were excluded because the data were 
not available in the source. NTSB's Performance and Accountability 
Report. Nonaviation accident investigations include highway, marine and 
rail, pipeline, and hazardous materials accident investigations. In 
fiscal year 2006, NTSB completed 19 nonaviation accident 
investigations. 

[End of figure] 

Both CSB�s and NTSB�s statutes allow the use of investigations 
performed by other agencies. Unlike CSB, NTSB uses its authority to 
solicit investigative work performed by others when resources or other 
considerations prevent it from deploying to accident sites. In all 
cases, overall investigative control, including determination of 
probable cause, rests with NTSB. 

NTSB's size prevents it from being on-site for many aviation 
investigations; therefore, the agency conducts limited investigations 
in which NTSB investigators do not go to the scene of the accident to 
gather information but rather correspond with local officials, rescue 
response units, Federal Aviation Administration personnel, and other 
persons and organizations that might have knowledge of the accident. 
The different types of NTSB aviation investigations are described in 
attachment 2. 

Similar to NTSB, CSB has the authority to use information gathered by 
local officials, rescue response units, Occupational Safety & Health 
Administration (OSHA) and EPA personnel, and other persons and 
organizations to prepare its reports and thereby expand its knowledge 
base of chemical accidents while operating with limited resources. 

CSB has memorandums of understanding with OSHA and EPA that indicate in 
certain instances, CSB may decide not to send an investigation team to 
the site of a chemical incident but may collect incident information 
from EPA, OSHA, or other on-site agencies compiled in the course of 
their own actions. 

Like NTSB, CSB has the ability to issue accident briefs, in addition to 
more detailed accident reports. See attachment 3 for a description of 
the types of NTSB accident reports. 

We reported in 2000 that CSB discontinued its limited, office-based 
review program that used investigative reports prepared by other 
organizations that respond to accidents. While the limited review 
program was less resource intensive than full investigations, board 
officials told us that they terminated this practice in 1999 because it 
conflicted with the board�s independence by having the board rely on 
the work of other agencies. 

In contrast, NTSB does conduct limited, office-based investigations 
that rely on the work of other entities. We reported in 2006 that while 
NTSB�s use of others� work may present some challenges, it appears to 
be working well. Performing data reliability assessments of important 
information based on the work of others is a generally accepted means 
used by federal and private audit organizations. CSB could likewise use 
existing reports and studies rather than performing comprehensive 
investigations of its own. 

[End of section] 

Data quality problems continue: 

CSB has not fully responded to IG recommendations to improve the 
quality of its accident data. 

Initial problem: CSB maintains an accident-screening database to report 
to Congress and the public on the number of chemical accidents it 
identifies and selects for investigation. However, the Department of 
Homeland Security (DHS) IG reported that CSB had inadequate control 
over the quality of its data and needed to monitor for completeness, 
accuracy, timeliness, and usefulness. 

The IG�s recommendation: The DHS IG recommended that CSB fulfill its 
statutory requirement to publish a regulation for receiving information 
from facilities on their chemical accidents. The IG also recommended 
that CSB develop a long-term strategy to address the shortfall in 
national chemical accident database quality. 

How CSB responded: CSB has not issued the regulation and has no plans 
to do so. According to CSB officials, the current system of monitoring 
media reports, searching the Web, and obtaining accident reports from 
NTSB and other sources is sufficient. 

Problem remaining: According to CSB officials, they have resolved the 
IG�s concerns about data quality; however, we found that CSB still has 
not published a data-reporting regulation and lacks a long-term 
strategy to improve quality controls, and the data remain somewhat 
inaccurate and incomplete. For example, when we analyzed a subset of 
accidents in the database involving fatalities and injuries, we found 
at least five accidents (about 6 percent) where fatalities were not 
correctly recorded in the database. We also found seven accidents 
(about 4 percent) where data on injuries were missing as a result of 
incomplete data entry. Moreover, CSB does not have procedures to ensure 
that data were entered accurately. 

CSB officials maintain that publishing a regulation to obtain 
information from facilities on their chemical accidents is not 
necessary because the information the board already receives is 
sufficient. Without a regulation, CSB relies primarily on the media, 
such as online newspapers and television, to learn about chemical 
accidents. It is likely that information reported by facilities would 
be a better source than CSB�s current practice of relying mostly on the 
media. Information provided in media sources may contain errors and not 
include the kind of information CSB needs to make decisions about which 
accidents to investigate or provide the agency with the kind of 
information needed for trend analysis and prevention outreach. 

In addition, the lack of quality controls of CSB�s accident database 
has contributed to data problems that could limit CSB�s ability to 
accurately report information on the investigative gap to Congress, 
target its resources, identify trends and patterns in chemical 
accidents, and prevent future similar accidents. 

[End of section] 

Human capital problems persist: 

CSB has not fully responded to recommendations to resolve its human 
capital problems. 

Initial problem: In 2002, the FEMA IG identified weaknesses in CSB�s 
human capital management, particularly that it had a shortfall of 
investigators, had not made hiring investigators a top priority, and 
that it lacked a central human capital manager, a comprehensive 
strategic human capital plan, and performance measures and criteria. 

IG�s recommendation: The FEMA IG recommended that CSB make hiring a top 
priority and made several recommendations to strengthen its human 
capital planning and management. 

How CSB responded: CSB consolidated human capital responsibilities 
under a full-time human resources manager, developed several agencywide 
strategies that included goals to improve human capital, hired more 
investigators, and established a remote office in Denver for one senior 
investigation manager. However, CSB officials acknowledged difficulties 
in attracting senior investigators to live in the District of Columbia 
and in retaining intern investigators. 

Problem remaining: CSB�s human capital strategy was not comprehensive, 
lacked a detailed action plan for closing the investigator shortfall, 
did not include input from staff investigators, and lacked performance 
measures�actions included in the Strategic Management of Human Capital 
portion of the President�s Management Agenda. Therefore: 

* CSB�s shortfall in investigators and problems retaining staff 
continue, limiting the number of accidents CSB could investigate 
regardless of resources. 

* According to a board member and some investigation managers and 
investigators we interviewed, intern investigators are encouraged to 
leave CSB in order to gain private industry experience or to pursue a 
terminal graduate degree. Encouraging interns to leave federal service 
for private industry experience does not follow the purpose of the 
Federal Career Intern program, which is designed to attract and retain 
employees for federal service. 

More employees left CSB in fiscal years 2006 and 2007 than were hired, 
indicating a possible problem with retention. 

CSB officials acknowledge a problem with hiring and retention; however, 
the agency has not given a retention bonus to an employee since 2002. 

Figure 2: Number of CSB Hires and Separations Fiscal Years 2006-2007: 

[See PDF for image] 

This figure is a stacked vertical bar graph depicting the following 
data: 

Number of employees hired: 
Noninvestigative staff: 7; 
Investigators: 4. 

Number of employees left: 
Noninvestigative staff: 9; 
Investigators: 5. 

Source: CSB's budget and staffing plan documents. 

[End of figure] 

In fiscal years 2006 and 2007, 3 of 5 investigators who left CSB were 
senior investigators with 5-7 years of experience. Yet CSB hired mostly 
interns during the same 2 fiscal years. 

In fiscal years 2006 and 2007, 4 of 11 staff hired were investigators. 
Three of the 4 investigators CSB hired were interns. CSB officials told 
us that they can hire interns more easily than senior investigators. 

Figure 3: CSB Investigators and Other Staff Hired, Fiscal Years 
2006�2007: 

[See PDF for image] 

This figure is a pie-chart depicting the following data: 

Investigators hired: 
Senior Investigator, GS-13: 1; 
Intern Investigators, GS-7: 3. 
Other staff: 7. 

Source: CSB's budget and staffing plan documents. 

[End of figure] 

According to CSB officials, an intern does not replace an experienced 
senior investigator; after converting to a permanent position from the 
intern position, it takes 3 to 5 years to become fully qualified. 

However, CSB officials said retaining interns is a challenge and noted 
that two of four intern investigators hired in fiscal years 2004 and 
2005 left CSB and one transferred to another office. 

If CSB had retained intern investigators hired in fiscal years 2004 and 
2005, it would have two additional investigators with 3 years of 
experience in fiscal year 2008. 

CSB has not fully used strategies available to federal agencies that 
are designed to recruit or retain staff, such as: 

* Federal Career Intern Program�The purpose of the federal intern 
program is to convert successful interns to permanent employees. The 
majority of interns have left CSB. 

* Education and training assistance�Intern investigators interested in 
pursuing a graduate degree might remain at CSB if the agency offered 
support for tuition and additional training. 

* Retention bonuses�Federal agencies may pay a retention incentive to a 
current employee if the agency determines that it is essential to 
retain the employee. CSB has not given any employee a retention bonus 
since September 2002. 

* Remote offices�CSB has not expanded the use of remote offices. 

In fiscal year 2006, CSB reprogrammed compensation funds of $627,891 to 
other priorities, such as safety videos and redesigning a Web site. In 
fiscal year 2007, CSB reprogrammed compensation funds of $407,383 to 
similar purchases. 

Had CSB made resolving human capital problems a top priority, these 
funds could have been used during the year for recruitment and 
retention bonuses, education and training assistance, and student loan 
repayment to hire and retain investigators. 

[End of section] 

Accountability and management problems continue: 

CSB has not delegated the authority to effectively manage the day-to-
day administrative functions to a permanent chief operating officer 
(COO) to ensure continuity of management and accountability. 

Initial problem: Between fiscal years 2000 and 2002, CSB did not have a 
permanent COO or a fully staffed board with a chair. Uncertainties 
surrounding the acting COO�s role and authority resulted in fractured 
management, a weakened chain of command, and board member intervention 
in routine administration. 

IG�s recommendation: The IG recommended that the board delegate to the 
COO the authority to effectively manage the day-to-day operations of 
CSB. 

How CSB responded: CSB hired a COO in June 2002, but the individual 
left after 2 years. The board subsequently eliminated the position and 
transferred the COO�s responsibilities to individual program managers 
and the board. 

* CSB board members said that a COO or similar executive director 
position might reduce the administrative responsibilities of the board, 
however, the CSB Chairman said a COO is unlikely to provide any 
additional management skills not already represented at the agency. Two 
investigation managers and two investigators we interviewed generally 
expressed support for reinstating the COO position to improve the 
continuity of administrative management. 

* EPA IG officials told us that there was support for the COO position 
among the CSB employees (former and current) they interviewed during 
their investigation in fiscal year 2007. These CSB employees indicated 
that they preferred how things operated when the COO was present. 

Problem remaining: CSB continues to operate without a permanent COO or 
a fully staffed board. Therefore: 

* CSB lacks a permanent senior executive with responsibility to 
establish performance goals, hold program managers accountable for 
meeting those goals, ensure there is a shared vision among board 
members and managers,and demonstrate improvement in the agency�s 
ability to meet it statutory mission over time. 

* CSB may not be able to ensure continuity of performance and 
accountability when board members and chairs leave the agency. 

* Regardless of the board�s staffing status, GAO has recommended in the 
past that employing a COO would provide long-term attention and focus 
on management issues. 

[End of section] 

IG oversight is still warranted: 

CSB�s history suggests that an IG�s continued independent institutional 
audit presence is the best option for ensuring that CSB is more 
accountable to Congress for meeting its public safety mission and 
statutory requirements.As we reported in 2000, three options for 
oversight include: 

1. establishing an in-house audit and investigations unit, 

2. contracting out for evaluations of its operations and programs, and, 

3. obtaining the services of an existing office of inspector general. 

We do not believe options 1 and 2 are appropriate for the board for 
several reasons. 

Option 1, an in-house audit unit, does not appear to be practical 
because CSB�s history of management problems warrants a level of 
independent oversight that may be difficult to achieve by an internal 
audit function. In addition, the limited staffing that would reasonably 
be allocated to this function at an agency of this size would lack the 
varied expertise need to address these problems. 

Option 2, contracting out for evaluations, also does not appear to be 
appropriate because of the limitations of contracting in terms of both 
audit independence and the potentially limited duration of the 
contracting relationship�which would not provide the continuity of 
oversight needed to address CSB�s long-standing management problems. 

CSB officials told us they prefer option 2 because IG oversight is 
inappropriate for an agency of CSB�s small size. CSB officials told us 
that CSB is currently classified as 1 of 54 federal entities,as defined 
by the IG Act of 1978, for which the act did not provide an IG. 

* Under the IG Act, federal entities are required to report annually to 
Congress and the Office of Management and Budget on whether an IG has 
been established or what other actions the federal entity took to 
otherwise ensure that audits of its programs and operations were 
conducted. 

CSB officials also told us that the IG Act provides a reasonable and 
responsible level of oversight for federal entities and requested that 
CSB be treated similarly to these entities. However, in our view, CSB�s 
long-standing, serious, and intractable management problems make it 
unlikely that this reporting requirement alone will ensure that the CSB 
has an appropriate level of oversight necessary to address its 
management problems. 

Given CSB�s management problems, the fact that other federal entities 
lack an IG should not determine whether CSB should have an IG. In any 
event, by statute some other federal entities do have another agency�s 
IG assigned to provide them with oversight. For example, the IG of the 
Department of Agriculture is assigned to provide oversight to the Delta 
Regional Authority. (Att. 1 provides the list of 54 federal entities, 
as defined by the IG Act of 1978, and identifies 4 that have a critical 
safety mission.) 

Option 3, obtaining the services of an existing IG office, appears to 
be the most appropriate oversight arrangement given the 
serious,intractable management problems at CSB. 

We believe that all significant federal programs and entities should be 
subject to oversight by IGs who can provide sound independent audits of 
all significant federal operations and activities. Given the management 
problems that our audit revealed, the need for independent IG oversight 
is especially pressing. 

While we believe that IG oversight remains the most appropriate 
oversight option, we found some shortcomings with the current EPA IG�s 
oversight relationship: 

* The current IG arrangement is not permanent, a fact that may 
undermine the continuity of oversight; the EPA IG is assigned to serve 
as CSB�s IG in annual appropriation bills. 

* EPA IG officials told us they have no plans to conduct program 
evaluations of CSB in fiscal year 2008 because they are allocating 
their limited evaluation resources to other priorities within EPA. 

However, we do not believe, as CSB asserts, that the EPA IG�s 
assignment and work call into question CSB�s intended independence from 
EPA. According to CSB, CSB independently evaluates and reports to 
Congress on EPA programs in chemical accident prevention and CSB�s 
independence from EPA was deliberate and carefully considered. By 
statute, inspectors general are independent from the agencies they 
audit and investigate so the EPA IG must maintain his or her 
independence from EPA officials and program employees. With such 
independence, the IG poses no risk to CSB�s independent evaluations of 
chemical accidents. 

[End of section] 

Conclusions: 

After 10 years of operation, CSB continues to operate in noncompliance 
with its statutory mandates. 

CSB stresses that it recognizes the importance of its investigations to 
identify root causes of accidental releases and recommend regulatory 
action to prevent such accidental releases but is not investigating all 
chemical accidents that have a fatality, serious injury, property 
damage, or the potential for a fatality, serious injury, or property 
damage. 

While we recognize that CSB may not have sufficient resources to 
investigate every accident within its purview, as NTSB reports it does, 
we believe CSB is missing opportunities to investigate more accidents 
and possibly prevent fatalities, serious injures, and substantial 
property damage in the future by not using the work of other entities. 

In addition, given the resource constraints on the board that limit its 
ability to investigate all chemical accidents resulting in fatalities, 
serious injuries, or substantial property damage, it is particularly 
important that CSB have the best available data on which to make 
decisions on those accidents that are most important to investigate. 
Even though CSB has a statutory requirement to issue a regulation 
requiring facilities to report their chemical releases, the board has 
resisted requiring such reporting, preferring to rely on alternative 
information sources, such as major news organizations. 

The difficulties that CSB has experienced in meeting its mission are 
largely the result of inadequate management accountability for 
addressing long-standing problems and for clearly identifying and 
attempting to meet CSB�s staff requirements to perform investigations 
of chemical accidents. In this regard, CSB has functioned without a 
permanent chief operations officer with responsibility for holding 
CSB�s managers accountable for their management activities and has not 
developed a comprehensive human capital strategy to improve its efforts 
to investigate chemical accidents in line with the board�s statutory 
responsibilities. 

[End of section] 

Recommendations to the Agency: 

We recommend that the chairman of the Chemical Safety Board: 

* develop a plan to address the investigative gap and request the 
necessary resources from Congress to meet CSB�s statutory mandate or 
seek an amendment to its statutory mandate; 

* consider using the work of other entities, such as government 
agencies, companies, and contractors (subject to an assessment of the 
quality of their work) to a greater extent to maximize the board�s 
limited resources; and; 

* improve the quality of its accident-screening database by better 
controlling data entry and periodically sampling accident data to 
evaluate their consistency and completeness. 

* Publish a regulation requiring facilities to report all chemical 
accidents, as required by law, to better inform the agency of important 
details about accidents that it may not receive from current sources. 

* Consider reinstating the position of chief operating officer, with 
the delegation of responsibility for establishing performance goals, 
holding program mangers accountable for meeting those goals, and 
demonstrating improvement in the agency�s ability to meet its statutory 
mandates over time. 

* Use the Strategic Management of Human Capital portion of the 
President�s Management Agenda to provide criteria for developing a 
comprehensive human capital plan, with input from investigators, that 
includes specific objectives and performance measures to improve 
accountability for results and to assist the agency in its goal of 
improving its human capital and infrastructure. 

[End of section] 

Matters for Congressional Consideration: 

Congress may wish to consider amending CSB�s organic statute or the 
Inspector General Act of 1978 to permanently give EPA�s Inspector 
General the authority to serve as the oversight body for the agency. 

As Congress prepares the appropriation of the EPA Inspector General, it 
may wish to consider providing the Inspector General with 
appropriations and staff allocations specifically for the audit 
function of CSB via a direct line in the EPA appropriation. 

[End of section] 

Attachment 1: Federal entities, as defined by the IG Act of 1978: 

1. Chemical Safety and Hazard Investigation Board*. 

2. Defense Nuclear Facilities Safety Board[A]. 

3. National Transportation Safety Board[A]. 

4. Nuclear Waste Technical Review Board[A]. 

5. Advisory Council on Historic Preservation. 

6. African Development Foundation. 

7. American Battle Monuments Commission. 

8. Architectural and Transportation Barriers Compliance Board. 

9. Armed Forces Retirement Home. 

10. Barry Goldwater Scholarship and Excellence in Education Foundation. 

11. Christopher Columbus Fellowship Foundation. 

12. Commission for the Preservation of America's Heritage Abroad. 

13. Commission of Fine Arts. 

14. Commission on Civil Rights. 

15. Committee for Purchase from People Who Are Blind or Severely 
Disabled. 

16. Court of Appeals for Veterans Claims. 

17. Court Services and Offender Supervision Agency for DC. 

18. Delta Regional Authority. 

19. Farm Credit System Insurance Corporation. 

20. Federal Financial Institutions Examination Council. 

21. Federal Mediation and Conciliation Service. 

22. Federal Mine Safety and Health Review Commission. 

23. Federal Retirement Thrift Investment Board. 

24. Harry S. Truman Scholarship Foundation. 

25. Inter-American Foundation. 

26. Institute of American Indian and Alaska Native Culture and Arts 
Development. 

27. Institute of Museum and Library Services. 

28. James Madison Memorial Fellowship Foundation. 

29. Japan-U.S. Friendship Commission. 

30. Marine Mammal Commission. 

31. Merit Systems Protection Board. 

32. Millennium Challenge Corporation. 

33. Morris K. Udall Scholarship and Excellence in National 
Environmental Policy Foundation. 

34. National Capital Planning Commission. 

35. National Commission on Libraries and Information Science. 

36. National Council on Disability. 

37. National Mediation Board. 

38. Neighborhood Reinvestment Corporation. 

39. Occupational Safety and Health Review Commission. 

40. Office of Government Ethics. 

41. Office of Navajo and Hopi Indian Relocation. 

42. Office of Special Counsel. 

43. Overseas Private Investment Corporation. 

44. Presidio Trust. 

45. Selective Service System. 

46. Smithsonian Institution/John F. Kennedy Center for the Performing 
Arts. 

47. Smithsonian Institution/National Gallery of Art. 

48. Smithsonian Institution/Woodrow Wilson International Center for 
Scholars. 

49. Trade and Development Agency. 

50. U.S. Holocaust Memorial Museum. 

51. U.S. Institute of Peace. 

52. U.S. Interagency Council on Homelessness. 

53. Vietnam Education Foundation. 

54. White House Commission on the National Moment of Remembrance. 

[A] Has a public safety mission. 

[End of attachment] 

Attachment 2: Types of NTSB Aviation Accident and Incident 
Investigations: 

Type of Investigations: Major investigation; 
Number in Fiscal Year 2006: 7; 
Description: Investigation of a significant accident that typically 
involves fatalities, multiple injuries, considerable property damage, 
or significant public interest. A team of investigators travels to the 
accident site. 
Travel involved? Yes. 
Fatalities involved? Yes. 

Type of Investigations: Field investigation[A]; 
Number in Fiscal Year 2006: 196; 
Description: Investigation of an accident that typically involves a 
fatality.At least one investigator travels and there is a significant 
amount of follow-up investigation from the office. 
Travel involved? Yes. 
Fatalities involved? Typically yes. 

Type of Investigations: Limited investigation[A]; 
Number in Fiscal Year 2006: 550; 
Description: Investigation of accidents that can involve a fatality but 
typically do not. Investigators do not travel to the accident site but 
instead conduct a limited investigation using information collected by 
Federal Aviation Administration officials, local officials, rescue 
response units and other persons, and organizations that may have 
knowledge of the incident. 
Travel involved? No. 
Fatalities involved? Typically no. 

Type of Investigations: Data collection accident investigation; 
Number in Fiscal Year 2006: 739. 
Description: Investigations of accidents that do not involve any 
fatalities, �critical� serious injuries, and other criteria. 
Investigators do not travel to the accident site but instead collect 
and analyze information from the office in order to determine the 
cause. 
Travel involved? No. 
Fatalities involved? No. 

Type of Investigations: Incident investigation[B]; 
Number in Fiscal Year 2006: 29; 
Description: Investigation of occurrences that do not involve 
fatalities. Investigators do not typically travel for these 
investigations but instead conduct an investigation from the office. 
Travel involved? No. 
Fatalities involved? Typically no. 

[A] Includes investigations of public use aircraft. 

[B] These occurrences do not meet criteria to be defined as an official 
accident but are still deemed important enough to investigate. 

[End of table] 

[End of attachment] 

Attachment 3: Types of NTSB Accident Reports: 

Type: Major accident reports; 
Description: Reports that provide detailed narrative accounts of the 
facts, conditions, circumstances, analysis, conclusions, and probable 
cause of an accident. This type of report is issued for major aviation, 
railroad, highway, pipeline, and marine accident investigations. 

Type: Nonmajor accident reports (or �accident briefs�); 
Description: Reports that briefly summarize the probable cause of an 
accident. This type of report is issued for all aviation accidents and 
for all nonmajor railroad, highway, pipeline, and marine accidents 
investigated by or for NTSB, for which probable cause is determined. 

[End of table] 

[End of attachment] 

[End of slide presentation] 

Enclosure III: Comments from CSB: 

Note: GAO comments supplementing those in the report text appear at the 
end of this enclosure. 

U.S. Chemical Safety and Hazard Investigation Board: 
2175 K Street, NW, Suite 650: 
Washington, DC 20037-1802: 
Phone: (202) 261-7600: 
Fax: 12021 261.7650: 
[hyperlink, http://www.csb.gov]: 

John S. Bresland, Chairman & CEO: 
Gary L. Visscher, Board Member: 
William B. Wark, Board Member: 
William E. Wright, Board Member: 

July 11, 2008: 

John B. Stephenson: 
Director: 
Natural Resources and Environment: 
U.S. Government Accountability Office: 
441 G Street, N.W. 
Washington, DC 20548: 

Dear Mr. Stephenson: 

As requested, we reviewed your draft report, "Chemical Safety Board: 
Improvements in Management and Oversight are Needed" that your office 
provided on June 11, 2008. We agree on various points in the draft 
report, such as the need to expand the investigation program. However, 
as you are aware, the Chemical Safety and Hazard investigation Board 
(CSB) has concerns with many of your conclusions and characterizations. 
Rather than reiterate our concerns, we will focus our comments on your 
recommendations. [See comment 1] 

Before discussing individual recommendations, however, there is one 
point that I particularly want to emphasize. The draft report refers to 
previous Inspector General (IG) recommendations, and leaves the 
impression, at least, that CSB has not been very responsive to those 
recommendations. In fact all previous IG recommendations have been 
closed by the respective IG's who made them. [See comment 2] 

Investigative Gap: 

GAO Recommendation: Develop a plan to address the investigative gap and 
request the necessary resources from the Congress to meet the CSB 's 
statutory mandate or seek an amendment to its statutory mandate. 

CSB Response: The CSB has not construed the agency's authorizing 
statute as requiring investigation of every chemical accident involving 
a fatality, serious injury, or substantial property damage, or the 
potential for such consequences, but understands GAO's concern and 
recommendation. We also agree that the performance of our mission - to 
help companies and communities avoid chemical accidents - would be 
strengthened by investigating and reporting on more of the serious 
chemical accidents that occur in the United States each year. [See 
comment 3] 

In addition to seeking additional investigation resources, we will 
draft a plan for obtaining information on additional chemical accidents 
occurring in the United States, and clearly set forth a risk-based 
approach to accident selection and investigation. We will also work 
with Congress to clarify the issue of CSB's statutory mandate, as 
suggested by GAO, including if appropriate, an amendment to the CSB's 
enabling legislation. [See comment 4] 

We will examine the work of the National Transportation Safety Board 
(NTSB) to aid our future plan. [See comment 5] However, we note that 
the GAO's comparison between 6 CSB investigations and 1,521 NTSB 
aviation investigations in FY 2006 may not be the most fruitful 
benchmark for future planning. The majority of the NTSB's aviation 
investigations were non-fatal general aviation incidents, with the NTSB 
relying heavily on the work of 3,400 FAA regional inspectors to 
determine the probable cause. Thus, the CSB will also examine the 
NTSB's statutory authority and work in other modes of transportation. 
In these other modes, the NTSB employs a risk based approach to select 
out priority incidents from all incidents within its jurisdiction. [See 
comment 6] 

The purpose of increased resources, the plan, legislative 
clarification, and other measures discussed below would be to provide 
greater oversight and analysis of significant chemical incidents in the 
United States. [See comment 7] 

Work of Others: 

GAO Recommendation: Consider using the work of other entities, such as 
government agencies, companies, and contractors (subject to an 
assessment of the quality of their work) to maximize the Board's 
limited resources. 

CSB Response: The Board will consider using the work of other entities 
and contractors to further maximize its limited resources. [See comment 
8] In our experience, however, there are limits and pitfalls to the use 
of other entities' work. Based on this experience, we respectfully 
suggest that the CSB has correctly interpreted its Congressional 
mandate by independently investigating major accidents and hazards in 
depth, rather than attempting to serve as a clearinghouse for numerous, 
disparate, and often superficial reports from other organizations. We 
discuss some of our initial concerns with GAO's recommendation below. 
[See comment 9] 

The Occupational Safety and Health Administration (OSHA) and the 
Environmental Protection Agency (EPA) frequently conduct inspections at 
accident sites. [See comment 10] However, they have few inspectors 
focused and specialized on chemical process safety, and these agencies 
typically do not prepare narrative reports describing what happened and 
why. Rather, OSHA and EPA may publish citations or notices of 
regulatory violations, which may or may not relate to the actual causes 
of the accidents. These lists of violations cannot generally be used to 
establish and report on the causes of the accidents, as required under 
the CSB's authorizing statute. Likewise, the investigations conducted 
by state and local police and fire agencies generally focus on 
particular inquiries such as the "cause and origin" of a chemical fire, 
i.e., a determination of the fuel, location, and ignition source. Such 
narrow determinations fall well short of Congress's stated intentions 
for the scope of CSB investigations. [Footnote 7] 

In addition, OSHA, EPA, state fire marshals, fire departments, and 
police clearly have law enforcement and regulatory responsibilities. 
Investigations by these agencies necessarily focus on rules violations, 
rather than on the overall adequacy of existing rules, standards, and 
industry practices. Because of their law enforcement and regulatory 
duties, many of these agencies are reluctant to share the results of 
their ongoing investigations with the CSB. In fact, during the CSB's 
investigation of the BP Texas City disaster, both OSHA and the EPA 
asserted in correspondence that they would limit the CSB's access to 
their inspection staff and records, citing in part the possibility of 
criminal prosecution. Today, more than three years after the BP 
accident - and more than a year after the CSB completed its landmark 
341-page report - issues of criminal responsibility are still being 
litigated in federal court. [See comment 11] 

Most companies that experience significant chemical accidents that 
cause deaths, injuries, or off-site consequences become involved in 
lengthy litigation proceedings. Even the most enlightened corporations 
generally produce their own accident investigation reports as part of 
their litigation defense strategy - not in an effort to inform other 
companies and the public objectively about underlying causes. Moreover, 
companies often assert that such reports are protected by legal 
privilege. Even if the CSB was able to obtain and rely on such reports 
as a primary source of information, it could undermine the credibility 
of the CSB's work and lead, in some cases, to incorrectly faulting the 
efforts of individual workers and third parties, rather than uncovering 
important systemic causes. [See comment 12] 

Quality of Accident Data: 

GAO Recommendation: Improve the quality of [the CSB 's] accident 
screening database by better controlling data entry and periodically 
sampling accident data to evaluate its consistency and completeness. 

CSB Response: We note that the screening database primarily represents 
a compilation of the earliest reports of accidents - including those 
from the National Response Center (NRC) and the media - which may 
contain inherent inaccuracies. Nonetheless the CSB agrees that it 
should take additional steps to prevent errors from being introduced 
through incorrect data entry. The CSB will revise its board order on 
the incident selection process and consider changes to improve data 
accuracy. We plan to consider such measures as additional written 
guidance and training for incident screeners; designing an electronic 
workflow so that significant changes to the database require 
supervisory sign-off; and periodic auditing of screening data for 
quality and completeness. In addition we will review the staffing for 
the screening program and its overall structure as the CSB develops its 
human capital plan. Currently screening is done by junior investigators 
as collateral duty on a rotating basis. The CSB lacks any staff solely 
dedicated to this task. 

Reporting Regulation: 

GAO Recommendation: Publish a regulation requiring facilities to report 
all chemical accidents, as required by law, to better inform the agency 
of important details about accidents that they may not receive from 
current sources. 

CSB Response: The CSB will publish in the Federal Register a Request 
for Information (RFI} concerning a reporting regulation. The RFI will 
present various options for rulemaking and seek the views and opinions 
of our stakeholders on the best path forward. We intend to publish the 
RFI within the next three months. In addition, the detailed plan to 
conduct more investigations will include staffing and resource 
projections for staff to collect and analyze incident information. [See 
comment 13] 

We note that the CSB's position has been that a reporting regulation is 
not needed for the narrow purpose of notifying the CSB of major 
accidents warranting the deployment of investigators, which appears to 
be the sole purpose of the CSB's authority to issue a reporting rule. 
[Footnote 8] Given the limited number of investigations that the CSB 
can conduct, we can and do easily learn what we need to know simply 
from monitoring the media and reports from the NRC and NTSB. [See 
comment 14] 

It is also important to note that the obligation to collect chemical 
incident data for the broader purposes identified by GAO appears to lie 
elsewhere in the federal government. For example, Congress funds 
programs to collect chemical accident data at both EPA and the Agency 
for Toxic Substances and Disease Registry. OSHA also collects chemical 
incident information. [See comment 15] 

Management Accountability and Continuity: 

GAO Recommendation: Consider reinstating the position of chief 
operating officer, with the delegations of responsibility for 
establishing performance goals, holding program managers accountable 
for meeting those goals, and demonstrating improvement in the agency's 
ability to meet its statutory mandates over time. 

CSB Response: The CSB agrees that it is appropriate to consider 
establishing a senior executive position to oversee important mission 
responsibilities. The CSB will give serious consideration to the 
establishment of such a position as part of its development of a 
strategic human capital plan. 

We note that GAO presented no evidence of a problem by not having a 
Chief Operating Officer (COO); rather they simply assert the CSB has 
"longstanding, serious, and intractable management problems." This is 
in stark contrast to the reality that the CSB's highest impact products 
(BP Texas City, Combustible Dust Study, CTA, Baker Panel, and the 
Safety Video Program) are from the post-2004 era when there was no COO. 
The GAO overlooks the fact that during the five years from 1998 to 
2002, the CSB completed a total of 12 investigations but then went on 
to complete 35 investigations from 2003 to 2008 - a near tripling of 
productivity on a flat budget. [See comment 16] 

It is also important to note that the CSB was without a Chairperson 
from August 2, 2007 to March 17, 2008, virtually the entire duration of 
the GAO's review. Consistent with its protocols, the Board delegated 
one Member interim executive and administrative authority to ensure 
orderly continuation of functions and duties, and there were no 
accountability or continuity issues identified during this period. 
Therefore, as the Board has evolved, the concern that the "CSB may not 
be able to ensure continuity of performance and accountability when 
Board Members and Chairs leave the agency," should be considered in the 
context of recent events which suggest otherwise. 

Human Capital Issues: 

GAO Recommendation: Use the Strategic Management of Human Capital 
portion of' the President's Management Agenda as criteria for 
developing a comprehensive human capital plan, with input from 
investigators that includes specific objectives and performance 
measures to improve accountability for results and to assist the agency 
in its goal of improving its human capital and infrastructure. 

CSB Response: The CSB agrees to use the Strategic Management of Human 
Capital portion of the President's Management Agenda (PMA) as a guide 
for developing a comprehensive human capital plan. [Footnote 9] The CSB 
will also continue to work with the Office of Personnel Management's 
(OPM) Small Agencies Human Capital Leadership and Merit System 
Accountability Office to develop the human capital plan. In addition, 
as recommended by OPM, the CSB will use the "Human Capital Report and 
Plan for the CSB's Office of Investigations" developed by the CSB human 
capital team in October of 2005 to help guide development of the 
comprehensive human capital plan. Developing the plan will be included 
in the CSB's FY 2009 action plan. 

Matters for Congressional Consideration: 

Finally, GAO's draft report raised two specific matters for 
Congressional consideration concerning oversight of the CSB. The CSB 
supports oversight and agrees that the CSB's authorizing legislation or 
the IG Act of 1978 are the proper places to address CSB oversight. 
However, the CSB has concerns with the GAO's specific suggestions for 
Congressional consideration. 

GAO Suggestion: Congress may wish to consider amending the CSB's 
authorizing statute or the Inspector General Act of 1978 to permanently 
give the EPA's Inspector General the authority to serve as the 
oversight body/or the agency. 

CSB Comment: The CSB respectfully disagrees with the GAO's suggestion 
that the EPA IG is the best, if not only, approach to provide permanent 
oversight for the CSB. While the EPA IG is one option for oversight of 
the CSB, other Offices of Inspector General are also available, some of 
which may be more appropriate for the role. [See comment 17] 

There are other alternatives as well. For example, the CSB is modeled 
on the NTSB. The NTSB, a much larger agency than the CSB, has never had 
and does not have its own Inspector General. Rather, GAO and the 
Department of Transportation Inspector General (DOTIG) share oversight 
responsibilities. GAO is responsible for broad oversight of the NTSB. 
[Footnote 10] DOTIG is Limited "to review only the financial 
management, property management, and business operations of the 
National Transportation Safety Board." (Emphasis added.) [Footnote 11] 
DOTTG actually contracts this function to a third party which it 
oversees, so the connection between DOTIG and the NTSB is extremely 
limited. [See comment 18] 

This model has been established in recognition of the fact that the 
NTSB issues recommendations to the Department of Transportation, 
similar to the manner in which the CSB issues recommendations to the 
EPA. Indeed, Congress has consistently and specifically rejected the 
assignment of DOTIG to serve as the Inspector General of the NTSB. 
[Footnote 12] The relevant legislative history explains that the 
limited nature of the DOTIG's authority was to "ensure that Inspector 
General oversight does not undermine the independence of the 
NTSB."[Footnote 13] [See comment 19] 

Another alternative is a small in-house Inspector General office. Such 
a model already exists for about 30 medium-size agencies, such as the 
Consumer Product Safety Commission, which are not large enough to have 
presidentially appointed IG's. Such an arrangement would appear to be 
less costly and more beneficial in the long term than what GAO has 
suggested. 

GAO Suggestion: As Congress prepares the appropriation of the EPA 
Inspector General, it may wish to consider providing the Inspector 
General with appropriations and staff allocations specifically for the 
audit function of CSB via a direct line in the budget. 

CSB Comment: The CSB is concerned that GAO did not adequately consider 
different oversight options for the CSB. While we agree that oversight 
of every federal agency is appropriate and can be beneficial, it should 
be tailored to the size of the agency. In that regard I note that the 
CSB currently obtains its financial and information security audits for 
about $60,000 a year. [See comment 21] 

Thank you for the opportunity to comment on your draft report. 

Sincerely, 

Signed by: 

John S. Bresland: 
Chairman & CEO: 

The following are GAO's comments to the U.S. Chemical Safety and Hazard 
Investigation Board's letter dated August 9, 2008. 

GAO Evaluation: 

1. We believe this briefing fairly and factually identifies continuing 
problems with CSB's governance, management, and oversight that have 
continued since we first reported on CSB, in fiscal year 2000.[Footnote 
14] We conducted this performance audit in accordance with generally 
accepted government auditing standards. Those standards require that we 
plan and perform the audit to obtain sufficient, appropriate evidence 
to provide a reasonable basis for our findings and conclusions based on 
our audit objectives. We believe that the evidence obtained provides a 
reasonable basis for our findings and conclusions based on our audit 
objectives. 

2. While the inspectors general closed these recommendations, their 
decision was often based on CSB's commitment to take future actions and 
the IG's reserving the right to reopen recommendations. As stated in 
the briefing, our analysis shows that CSB has not fully responded to 
key recommendations related to (1) investigating more accidents that 
meet statutory requirements triggering CSB's responsibility to 
investigate, (2) improving the quality of its accident data, (3) 
resolving human capital problems, and (4) ensuring accountability and 
continuity of management. 

3. The 1990 Clean Air Act Amendments establishing CSB contained the 
following mandates: 

* "The Board shall investigate (or cause to be investigated)�any 
accidental release resulting in a fatality, serious injury or 
substantial property damage"[Footnote 15] (emphasis added). 

* "In no event shall the Board forgo an investigation where an 
accidental release causes a fatality or serious injury among the 
general public, or had [sic] the potential to cause substantial 
property damage or a number of deaths or injuries among the general 
public"[Footnote 16] (emphasis added). 

As noted in our briefing slides, this language clearly identifies the 
accidental releases CSB is required to investigate; investigations are 
required for all accidental releases that result, or have the potential 
to result, in a fatality, serious injury, or substantial property 
damage. 

4. While CSB reports that it will seek additional investigation 
resources and "will draft a plan for obtaining information on 
additional chemical accidents occurring in the United States, and 
clearly set forth a risk-based approach to accident selection and 
investigation," it does not commit to meet CSB's statutory mandate or 
commit to investigating more than six accidents per year. However, CSB 
reported that the agency would work with Congress to clarify the issue 
of its statutory mandate and, if appropriate, seek an amendment to its 
authorizing statute. 

Also see comment 3. 

5. A more useful benchmark for comparison between CSB and NTSB is to 
examine their appropriations and accident investigations. As noted on 
slide 15, while NTSB's budget is 8 times CSB's budget, NTSB 
investigated 250 times as many accidents. In addition, while CSB notes 
that the majority of NTSB's accident investigations were nonfatal, the 
majority of accidents CSB screened were also nonfatal, as stated on 
slide 14. 

CSB also notes that NTSB relies "heavily on the work of FAA." CSB could 
similarly rely more on the work of EPA and OSHA and already has 
cooperative agreements in place with these agencies to share 
information on accidents. The agency currently has memorandums of 
understanding (MOU) with both EPA and OSHA with the stated purpose of 
establishing "policy and general procedures for cooperation and 
coordination between the two Agencies to minimize duplication of 
activities." Moreover, the MOUs state the two agencies will "coordinate 
incident notification, data and information exchange, training, 
technical and professional assistance." The MOU with EPA further states 
that "in certain instances, the CSB may decide not to send an 
investigation team to the site of a chemical incident. Rather, CSB may 
collect incident information from EPA or other on-site agencies 
compiled in the course of their own actions." Similar language appears 
in the MOU with OSHA. 

6. In our view, this approach would not be consistent with CSB's own 
authorizing legislation. NTSB's authorizing legislation gives it broad 
discretion in determining which accidents in the nonaviation 
transportation modes to investigate. For example, NTSB "shall 
investigate or have investigated�a highway accident, including a 
railroad grade crossing accident, the Board selects in cooperation with 
a State[Footnote 17]" (emphasis added). In addition, NTSB's authorizing 
legislation requires investigations of "any other accident related to 
the transportation of individuals or property when the Board decides" 
that certain circumstances are present (emphasis added[Footnote 18]). 
CSB's authorizing statute, on the other hand, does not contain any 
similar language providing discretion for CSB to determine which 
accidental releases to investigate. 

CSB's statutory authority is more comparable to the language that 
governs NTSB's investigations of civil aviation, railroad, and pipeline 
accidents. NTSB is required to investigate railroad accidents with a 
fatality or substantial property, railroad accidents that involve a 
passenger train, and pipeline accidents with fatalities, substantial 
property damage, or significant injury to the environment.[Footnote 19] 
Like CSB, NTSB is required to investigate these types of accidents; it 
does not have the discretion to choose which to investigate. 

Also see comment 3 and 4. 

7. Our recommendation is to develop a plan to address the investigative 
gap and request the necessary resources from Congress to meet CSB's 
statutory mandate or seek an amendment to its statutory mandate. 

Also see comment 3 and 6. 

8. GAO agrees that there are limitations in using the work of others. 
As CSB examines NTSB's work, it may find that there are lessons to be 
learned about how to minimize the limitations in using the work of 
others. While NTSB uses its authority to solicit the investigative work 
performed by others when resources or other considerations prevent it 
from deploying to accident sites, in all cases, overall investigative 
control--including determination of probable cause--rests with NTSB. As 
we reported in 2007, outside experts provide critical assistance to 
NTSB investigators. During the course of an investigation, NTSB 
supplements its investigative staff through the use of "parties" and 
outside contractors when it needs additional support for fact finding 
or technical analysis. "Party" participants include individuals, 
agencies, companies, and associations that can provide technical 
expertise relevant to a specific accident during the fact-finding 
phase. While the party process may provide technical information that 
is important for determining the cause of an accident, it presents 
inherent conflicts of interest for entities that are both parties in an 
investigation and potential defendants in related litigation. For 
example, in a commercial aviation accident, the principal defendants in 
litigation for damages are likely to be the airline and aircraft 
manufacturer, who may face liability for dozens of deaths--both 
entities that are likely working with NTSB as parties to the 
investigation. Despite such challenges, the party system appears to be 
working well; for example, RAND Corporation found that the party system 
works well in that it allows NTSB to leverage its resources to provide 
critical safety information in regard to the accident under 
investigation. In addition, NTSB officials told us that the system is 
an efficient way of gathering and sharing information about the 
accident with investigators and other parties. Also, having multiple 
parties on an investigation offsets concerns of conflict of interest 
and impartiality. 

In addition, according to NTSB officials, the agency makes a 
distinction between fact gathering and analysis. Parties are permitted 
to gather facts (evidence) for NTSB, but they are not permitted to 
engage in the analysis of that evidence. 

9. As noted above, CSB's authorizing statute requires investigations of 
all accidental releases that cause, or have the potential to cause, a 
fatality, serious injury, or substantial property damage. CSB's 
authorizing statute does not specify, that investigations should be 
limited to those accidental releases that are "major." In addition, 
CSB's authorizing statute grants CSB discretion to "utilize the 
expertise and experience of other agencies."[Footnote 20] 

Also see comment 3. 

10. CSB could use information available from other entities, including 
OSHA and EPA, to conduct its own independent analysis of accidents, 
although in some cases, CSB may need to supplement information provided 
by other entities through follow-up calls, sending investigators to the 
scene of the accident, or other means. 

Also see comment 8. 

11. When CSB examines the work of NTSB, the agency could review how 
NTSB obtains information from other public entities pertaining to 
ongoing investigations. 

Also, see comments 5 and 8. 

12. An examination of NTSB's work may provide information about how 
NTSB uses its party process to minimize the limitations in using the 
work of others. 

Also see comments 8 and 11. 

13. We recognize that obtaining the views and opinions of CSB's 
stakeholders could provide valuable information regarding the 
preparation of a reporting regulation. However, the request for 
information does not in itself provide assurance that CSB will follow 
through and issue a regulation as required by CSB's authorizing 
statute. Specifically, the statute provides that the board 
"shall�establish by regulation requirements binding on persons for 
reporting accidental releases into the ambient air subject to the 
Board's investigatory jurisdiction."[Footnote 21] CSB's comments 
concerning the need for such a regulation are not relevant because CSB 
is legally required to promulgate a regulation. In addition, we 
disagree with CSB about the regulation's usefulness. A reporting 
regulation would allow CSB to obtain more accurate, complete 
information to meet its statutory mandate. 

14. As we note in our briefing, information reported by facilities 
would be a better source than CSB's current practice of relying mostly 
on information from the media, which CSB officials acknowledge is often 
inaccurate or incomplete. Our analysis of CSB's accident data showed 
that in fiscal year 2007, CSB received 66 percent of its chemical 
accident notifications exclusively from the news media. In addition, 
CSB's accident database is not just for selection of accidents to 
investigate, it is also used to report to Congress and serves as a 
historical record of information on chemical accidents that could be 
used to identify trends and patterns in chemical accidents and prevent 
future accidents. 

See also comment 13. 

15. The Department of Homeland Security Inspector General (DHS IG) 
reported in 2004 that the lack of comprehensive and timely reporting on 
chemical accidents in general is a problem the CSB is both positioned 
and required to address.[Footnote 22] Citing a 2002 CSB report on 
reactive hazards,[Footnote 23] the DHS IG reported that "CSB searched 
40 public and private databases and reported that its findings were 
limited because existing sources of incident data are inadequate to 
identify the number, severity, frequency, and causes of reactive 
incidents. CSB's analysis included chemical accident data at EPA and 
OSHA. 

Also see comments 13 and 14. 

16. We introduced the recommendation for reinstating a permanent COO as 
one way CSB could begin addressing the management and accountability 
problems that we have presented in this briefing. As stated in the 
briefing, the difficulties that CSB has experienced in meeting its 
mission are largely the result of inadequate management accountability 
for addressing long-standing problems. If CSB officials recognize these 
long-standing problems and agree that they need to be addressed, then 
reinstating a permanent COO could help provide the continuity of 
management needed to bring about this change. In addition, former 
Chairman Carolyn W. Merritt testified in July 2007 that CSB: 

"would benefit�if Congress�provided for a vice chairman to assure the 
orderly transition during times when the chair is vacant. Periodic 
vacancies in the chair, and the resulting absence of executive 
authority, pose a significant risk to the success of the agency. Under 
the existing structure, CSB board members cannot serve beyond the 
expiration of their five-year terms, and thus vacancies in the chair 
and other board seats are all but inevitable." [Footnote 24] 

17. While we recognize that alternatives exist for providing CSB with 
management oversight, after reviewing various oversight options, we 
continue to believe that the EPA IG has several advantages, provides 
the best option for oversight, and the oversight authority should be 
made permanent. The EPA IG has expertise involving the chemical 
management issues that the Board is charged with investigating, has 
gained knowledge of CSB's operations and activities in providing the 
Board with oversight over the past several years, and, like other IGs, 
has the requisite independence provided by the IG Act of 1978 necessary 
for reviewing and making recommendations to address long-standing 
problems in the Board's management performance. 

18. We recognize that Congress provided for a division of 
responsibility between the Department of Transportation (DoT) IG and 
GAO regarding the oversight of NTSB. However, GAO has largely devoted 
its efforts to program evaluations and policy analyses that look at 
programs and functions across government and with a longer-term 
perspective. On the other hand, IGs have been on the front line of 
combating fraud, waste, and abuse within their respective agencies, and 
their work has generally concentrated on specific program-related 
issues of immediate concern with more of their resources going into 
uncovering inappropriate activities and expenditures through an 
emphasis on investigations. IGs play a critical role in identifying 
mismanagement of scarce taxpayer dollars and could provide, based on 
the results of our evaluation, a valuable service for the CSB. 
Moreover, as we have stated in the past, we believe that all 
significant federal programs and entities should be subject to 
oversight by independent IGs. 

See comment 17. 

19. In response to CSB's point that Congress has "consistently and 
specifically rejected" assigning the DoT IG to serve as IG of the NTSB, 
we note that the DoT IG's assignment to and GAO annual audit 
requirements of the NTSB are recent actions--the DoT IG was assigned to 
NTSB in 2000 and GAO assigned in late 2006. As stated in the briefing, 
we believe that all significant federal programs and entities should be 
subject to oversight by IGs who can provide sound, independent audits 
of all significant federal operations and activities. While we 
recognize that the Congress took specific actions it felt appropriate 
to ensure the independence of NTSB, we note that Congress has not taken 
such action with respect to CSB. Furthermore, IGs are independent from 
the agency they audit and investigate, and we believe the EPA IG has 
the requisite independence from CSB. 

See comment 17 and 18. 

20. As stated in the briefing, CSB's history of management problems 
warrants a level of independent oversight that may be difficult to 
achieve by an internal audit function. In addition, the limited 
staffing that would reasonably be allocated to this function at an 
agency of this size would lack the varied expertise needed to address 
these problems. 

21. As stated in the briefing, we believe that all significant federal 
programs and entities should be subject to oversight by IGs who can 
provide sound, independent audits of all significant federal operations 
and activities. Given the management problems that our audit revealed, 
the need for independent IG oversight is especially pressing. We also 
note that these management problems would not be addressed by financial 
or information security audits. 

See comment 17, 19 and 20. 

[End of enclosure] 

Footnotes: 

[1] S. Rpt. No. 101-228, 1990 U.S.C.C.A.N. 3385, 3615 (1989). 

[2] NTSB is required by statute to investigate every civil aviation 
accident in the United States and certain railroad, pipeline, and 
marine accidents and issue safety recommendations aimed at preventing 
future accidents. 

[3] GAO, Chemical Safety Board: Improved Policies and Additional 
Oversight Are Needed, [hyperlink, http://www.gao.gov/cgi-
bin/getrpt?GAO/RCED-00-192] (Washington, D.C.: July 11, 2000). 

[4] Joint Explanatory Statement to Accompany the Consolidated 
Appropriations Amendment, Division F, at 60. 

[5] Pub. L. No. 95-452, 92 Stat. 1101 (codified as amended at 5 U.S.C. 
App. 3). 

[6] GAO, National Transportation Safety Board: Progress Made, 
Management Practices, Investigation Priorities, and Training Center Use 
Should Be Improved, [hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-
07-118] (Washington, D.C.: Nov., 22, 2006). 

[7] See S. Rpt. No. 101-228, 1990 U.S.C.C.A.N. 3385, 3617-18 (1989) 
(stating that, "The Board should take an "all cause" theory....it is 
expected that the Board will follow many strands of inquiry...."). 

[8] See 42 U.S.C. � 7412(r)(6)(C)(iii). 

[9] OPM indicated to the CSB that it is not a scored agency under the 
PMA and therefore not required to go to the depth a scored agency mast. 
However, the CSB will develop the human capital plan using the PMA as a 
guide. 

[10] See Pub, L. No. 109-443, 120 Stat. 3297, 3299, � 5(a) (2006) 
(codified at 49 U.S.C.A. � 1138). 

[11] 49 U.S.C. � 1137(a). 

[12] The CSB has conducted numerous discussions with NTSB officials in 
order to understand why the DOTIG role as the NTSB IG was rejected. 
NTSB officials have consistently explained to us that this was to 
prevent the appearance or the occurrence of conflicts with DOT 
officials to whom the NTSB makes recommendations. 

[13] HR. Rep. No. 106-335 at 9 (2000) and S. Rep. 106-386 at 9 (2000). 

[14] GAO, Chemical Safety Board: Improved Policies and Additional 
Oversight Are Needed (GAO/RCED-00-192, July 11, 2000). 

[15] 42 U.S.C. � 7412(r)(6)(C)(i). 

[16] 42 U.S.C. � 7412(r)(6)(E). 

[17] 49 U.S.C. � 1131(a)(1)(B). 

[18] 49 U.S.C. � 1131(a)(1)(F). 

[19] 42 U.S.C. � 1131(a)(1). 

[20] 42 U.S.C. � 7412(r)(6)(D). 

[21] 42 U.S.C. � 7412(r)(6)(C)(iii). 

[22] Department of Homeland Security Office of Inspector General, A 
Report on the Continuing Development of the U.S. Chemical Safety and 
Hazard Investigation Board, OIG-04-04 (Jan. 7, 2004). 

[23] U.S. Chemical Safety and Hazard Investigation Board, Improving 
Reactive Hazard Management, Sept. 17, 2002. 

[24] Testimony of Carolyn W. Merritt, Chairman and Chief Executive 
Officer, U.S. Chemical Safety and Hazard Investigation Board before the 
U.S. Senate Committee on Environment and Public Works, Subcommittee on 
Transportation Safety, Infrastructure Security, and Water Quality, July 
10, 2007. 

[End of section] 

GAO's Mission: 

The Government Accountability Office, the audit, evaluation and 
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 GAO's Web site [hyperlink, http://www.gao.gov]. Each 
weekday, GAO posts newly released reports, testimony, and 
correspondence on its Web site. To have GAO e-mail you a list of newly 
posted products every afternoon, go to [hyperlink, http://www.gao.gov] 
and select "E-mail Updates." 

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: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: 
E-mail: fraudnet@gao.gov: 
Automated answering system: (800) 424-5454 or (202) 512-7470: 

Congressional Relations: 

Ralph Dawn, Managing Director, dawnr@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, D.C. 20548: 

Public Affairs: 

Chuck Young, Managing Director, youngc1@gao.gov: 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, D.C. 20548: