Superfund: Extent to Which Most Reforms Have Improved the Program Is
Unknown (Letter Report, 05/12/2000, GAO/RCED-00-118).

Pursuant to a congressional request, GAO reviewed the Environmental
Protection Agency's (EPA) Superfund program administrative reforms,
focusing on the: (1) reforms' demonstrated results and the performance
measures EPA uses to gauge these results; and (2) legislative changes to
the program that either EPA or key stakeholders--including, among
others, officials representing parties responsible for cleanups,
environmental groups, and states--believe are still necessary.

GAO noted that: (1) EPA claims and stakeholders agree that, in general,
the Superfund program has improved and the administrative reforms have
collectively contributed to this improvement; (2) however, GAO
determined that, for a majority of the 62 reforms, it is difficult for
EPA to demonstrate the extent to which they are working and have met the
goals set for them--to make the program faster, fairer, and more
efficient; (3) while maintaining that all the reforms are important, EPA
reform managers acknowledged that 42 reforms did not have a fundamental
effect, and EPA could not easily collect the data to measure the results
achieved for most of them; (4) 20 reforms had a fundamental effect, and
for these reforms: (a) EPA's performance measures demonstrated that 7
had achieved benefits, such as dollar savings--EPA has saved $70 million
to date by identifying less costly cleanup alternatives--and greater
community involvement in cleanups; (b) EPA's measures counted the number
of times that 7 were implemented but did not demonstrate the results
achieved; and (c) EPA did not have measures to demonstrate the results
that 6 had achieved; (5) EPA's data for the 14 fundamental and
measurable reforms show two trends suggesting that the progress made to
date may be eroding; (6) the implementation rates for almost half of
these reforms peaked in fiscal year 1997 and declined in subsequent
years; (7) the implementation rates for some reforms varied widely among
the regions, possibly indicating inconsistent application; (8)
stakeholders identified regional inconsistency as a problem with some
reforms; (9) therefore, better measurement and oversight of the key
reforms, as well as better understanding of the reasons for regional
variation in the implementation of some, could help EPA obtain the
maximum benefits possible from its reform initiative; (10) EPA and
stakeholders agree that targeted legislative changes would do more than
EPA's administrative reforms to protect certain parties from the current
Superfund law's liability provisions, however, they disagree on the
extent of change; (11) according to EPA, it is not seeking any
legislation to codify its reforms, but it would support legislative
proposals to limit liability for some parties that stakeholders have
identified; (12) these parties include prospective purchasers of
contaminated property and owners who are not responsible for or aware of
contamination on their property; and (13) EPA does not see a need for
other legislative changes, such as limiting liability for small
businesses, because it believes its reforms have created the tools
needed to provide relief for these parties.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  RCED-00-118
     TITLE:  Superfund: Extent to Which Most Reforms Have Improved the
	     Program Is Unknown
      DATE:  05/12/2000
   SUBJECT:  Waste disposal
	     Hazardous substances
	     Performance measures
	     Environmental legislation
	     Cost effectiveness analysis
	     Pollution control
	     Liability (legal)
	     Cost control
	     Environmental monitoring
	     Program evaluation
IDENTIFIER:  Superfund Program

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************

GAO/RCED-00-118

Appendix I: Fourteen Reforms That EPA Considers as Fundamental
and as Having Achieved Measurable Results

24

Appendix II: Characteristics of 48 Reforms That EPA Did Not
Classify as Both Fundamental and Measurable

78

Appendix III: Objectives, Scope, and Methodology

110

Appendix IV: Comments From the Environmental Protection
Agency

113

Appendix V: GAO Contacts and Staff Acknowledgments

117

Table 1: Fourteen Fundamental/Measurable Reforms and Their Output and
Outcome Measures 10

Table 2: Performance Goals and Measures for Encouraging the Use of
Alternative Dispute Resolution 24

Table 3: Performance Goals and Measures for Promoting "Enforcement First" 28

Table 4: Performance Goals and Measures for Promoting Settlements With Small
Waste Contributors 31

Table 5: Performance Goals and Measures for Reaching Agreements With
Prospective Purchasers 35

Table 6: Performance Goals and Measures for Providing Compensation for
Orphan Shares 39

Table 7: Performance Goals and Measures for Encouraging the Use of
Site-Specific Accounts 43

Table 8: Performance Goals and Measures for Revised Guidance on Settlements
With Miniscule Waste Contributors 47

Table 9: Performance Goals and Measures for Updating Remedy Decisions 49

Table 10: Performance Goals and Measures for Increasing Construction
Completions 53

Table 11: Performance Goals and Measures for Establishing the National
Remedy Review Board 57

Table 12: Number of Board Reviews and Estimated Savings, Fiscal Years
1996-2000 58

Table 13: Performance Goals and Measures for Using the Superfund Accelerated
Cleanup Model 61

Table 14: Performance Goals and Measures for Promoting Community Involvement
67

Table 15: Performance Goals and Measures for Funding Brownfield Assessment
Pilot Projects 74

Table 16: Six Fundamental Reforms That Lack Performance Measures 79

Table 17: List of 42 Reforms That EPA Did Not Specifically Identify as
Having Fundamentally Changed the Superfund Program 88

Figure 1: Comparison of Four Regions' Superfund Workloads and Rates of
Implementation for Three Reforms 15

Figure 2: Superfund Liability Cases Initiated Using Alternative Dispute
Resolution, Fiscal Years 1993-98 25

Figure 3: Each Region's Percentage of National Superfund Cases in Which
Alternative Dispute Resolution Was Used, Fiscal Years 1993-98 26

Figure 4: Percentage of New Cleanup Actions Funded by Responsible Parties,
Fiscal Years 1987-99 29

Figure 5: Average Percentage of New Cleanups Funded by Responsible Parties,
by Region, Compared With the National Average, Fiscal Years 1995-99 30

Figure 6: Number of De Minimis Settlements, Fiscal Years 1987-99 32

Figure 7: Each Region's Percentage of National De Minimis Settlements,
Fiscal Years 1987-99 33

Figure 8: Number of Prospective Purchaser Agreements Completed, Fiscal Years
1989-99 36

Figure 9: Each Region's Percentage of National Prospective Purchaser
Agreements, Fiscal Years 1989-99 37

Figure 10: Number of Orphan Share Compensation Offers, Fiscal Years 1996-99
40

Figure 11: Each Region's Percentage of National Orphan Share Compensation
Offers, Fiscal Years 1996-99 41

Figure 12: Number of Special Accounts Established, Fiscal Years 1990-99 44

Figure 13: Each Region's Percentage of National Special Accounts
Established, Fiscal Years 1990-99, Compared With Its Percentage of National
Superfund Sites 45

Figure 14: Number of Cleanup Remedy Updates per Year, Fiscal Years 1982-99
50

Figure 15: Each Region's Percentage of National Cleanup Remedy Updates,
Fiscal Years 1996-99, Compared With Its Percentage of National Superfund
Sites 51

Figure 16: Number of Sites Designated as Construction Complete, Fiscal Years
1981-99 54

Figure 17: Each Region's Percentage of National Construction Complete Sites,
Fiscal Years 1981-99, Compared With Its Percentage of National Superfund
Sites 55

Figure 18: Each Region's Percentage of National Remedy Review Board Reviews
58

Figure 19: Number of Non-Time-Critical Removals, Fiscal Years 1980-99 62

Figure 20: Each Region's Percentage of National Non-Time-Critical Removals,
Fiscal Years 1992-99, Compared With Its Percentage of National Superfund
Sites 63

Figure 21: Each Region's Percentage of National Integrated Assessments,
Fiscal Years 1994-99, Compared With Its Percentage of National Superfund
Sites 64

Figure 22: Community Advisory Groups Established, Through the End of Fiscal
Year 1999 68

Figure 23: Technical Assistance Grants Awarded, Fiscal Years 1988-99 69

Figure 24: Technical Outreach Projects Initiated, Fiscal Years 1994-99 70

Figure 25: Each Region's Percentage of National Community Involvement
Activities, Through the End of Fiscal Year 1999 71

CERCLA Comprehensive Emergency Response, Compensation, and Liability Act

EPA Environmental Protection Agency

GAO General Accounting Office

GPRA Government Performance and Results Act

NPL National Priorities List

OMB Office of Management and Budget

Resources, Community, and
Economic Development Division

B-284890

May 12, 2000

Congressional Requesters

For years, the Environmental Protection Agency's (EPA) Superfund program has
been criticized for a number of problems. These have included the pace and
cost of cleaning up hazardous waste sites, the agency's approach for holding
waste contributors liable for cleaning up sites, and the overall
effectiveness of the program. Responding to these criticisms, in June 1993,
EPA began implementing a series of 62 administrative reforms--actions that
it could take under its existing legal authority to improve the program's
fairness, efficiency, and effectiveness. According to EPA, these reforms
have led to faster cleanups of more sites, thereby better protecting public
health and the environment. In addition, EPA maintains that the reforms have
promoted the selection of more cost-effective cleanup methods and fairer
enforcement of the law. The agency has publicly stated that, because of the
administrative reforms, comprehensive legislative reform of the Superfund
program is no longer necessary, although the agency is willing to support
certain targeted legislative changes.

As the Congress considers reauthorizing the Superfund law, it has been
debating the extent to which legislative changes to the program would be
desirable. A key factor in making this determination is the extent to which
EPA's administrative reforms have addressed previously identified weaknesses
in the program. In this context, you requested that we review EPA's reforms
to

� determine their demonstrated results and evaluate the performance measures
the agency uses to gauge these results and

� identify legislative changes to the program that either the agency or key
stakeholders--including, among others, officials representing parties
responsible for cleanups, environmental groups, and states--believe are
still necessary.

We obtained basic information from EPA for all 62 reforms, including their
characteristics, scope, implementation status, overall goals, and
performance measures, where available. We also conducted a more detailed
analysis of a subset of 14 reforms that the agency characterized as having
significantly and measurably improved the program. Appendix I provides a
summary of our analysis of each of the 14 reforms. Appendix II provides a
summary of the information we collected on the remaining 48 reforms. To
determine the results of the reforms and any legislative changes needed, we
met with a judgmental sample of officials representing key stakeholders
affected by the Superfund program, including various industry groups, state
cleanup agencies, and environmental groups. Officials representing industry
and state cleanup agencies provided the majority of the comments about the
reforms. (App. III provides a listing of the stakeholders we contacted and a
more detailed discussion of our scope and methodology.)

EPA claims and stakeholders agree that, in general, the Superfund program
has improved and the administrative reforms have collectively contributed to
this improvement. However, we determined that, for a majority of the 62
reforms, it is difficult for the agency to demonstrate the extent to which
they are working and have met the goals set for them--to make the program
faster, fairer, and more efficient. While maintaining that all the reforms
are important, EPA reform managers acknowledged that

� 42 reforms did not have a fundamental effect, and EPA could not easily
collect the data to measure the results achieved for most of them;

� 20 reforms had a fundamental effect; for these reforms,

� the agency's performance measures demonstrated that 7 had achieved
benefits, such as dollar savings--EPA has saved $70 million to date by
identifying less costly cleanup alternatives--and greater community
involvement in cleanups;

� the agency's measures counted the number of times that 7 were implemented
but did not demonstrate the results achieved; and

� the agency did not have measures to demonstrate the results that 6 had
achieved.

Furthermore, EPA's data for the 14 fundamental and measurable reforms show
two trends suggesting that the progress made to date may be eroding. First,
the implementation rates for almost half of these reforms peaked in fiscal
year 1997 and declined in subsequent years. Second, the implementation rates
for some reforms varied widely among the regions, possibly indicating
inconsistent application. Moreover, stakeholders identified regional
inconsistency as a problem with some reforms, and EPA acknowledged that
ensuring such consistency is a challenge. Therefore, better measurement and
oversight of the key reforms, as well as better understanding of the reasons
for regional variation in the implementation of some, could help EPA obtain
the maximum benefits possible from its reform initiative. We are making
recommendations that the agency take such actions.

EPA and stakeholders agree that targeted legislative changes would do more
than the agency's administrative reforms to protect certain parties from the
current Superfund law's liability provisions; however, they disagree on the
extent of change. According to EPA, it is not seeking any legislation to
codify its reforms, but it would support legislative proposals to limit
liability for some parties that stakeholders have identified. These parties
include prospective purchasers of contaminated property and current owners
who are not responsible for or aware of contamination on their property. EPA
does not see a need for other legislative changes, such as limiting
liability for small businesses, because it believes its reforms have created
the tools needed to provide relief for these parties.

In 1980, the Congress passed the Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA), commonly known as Superfund, to
clean up highly contaminated hazardous waste sites. EPA places sites that it
determines may need long-term cleanup actions, called remedial actions, on
its National Priorities List (NPL). For sites needing cleanup, EPA or
private parties conduct studies to assess the risks and select, design, and
construct cleanup remedies. CERCLA authorizes EPA to compel the parties
responsible for the contaminated sites to clean them up. Under CERCLA, any
responsible party at a site can, under some circumstances, be held
responsible for the entire cost of the cleanup.1 Responsible parties can, in
turn, sue other parties to recoup some of their own expenses. Through this
process, parties can incur high legal costs. The law also allows EPA to pay
for cleanups and seek reimbursement from the parties, and it established a
trust fund, financed primarily by taxes on crude oil and chemicals, to help
EPA pay for its cleanups and related activities. The Superfund program's
authorization and the authority for the taxes financing the trust fund
expired in 1995 and have not been renewed. The Congress continues to fund
the program through annual appropriations from the Superfund trust fund and
general revenues.

Meanwhile, to address concerns about the high costs and long duration of
cleanups, EPA, beginning in 1993, publicly announced three separate rounds
of administrative reforms--actions it could take within its existing
legislative authority. These include

� 17 reforms announced in June 1993,

� 19 reforms announced in February 1995, and

� 26 reforms announced in October 1995.

These 62 reforms were intended to cover a range of activities, such as (1)
providing liability relief to certain parties, including contributors of
small volumes of waste or innocent landowners and purchasers, (2) selecting
more technologically advanced and cost-effective cleanup remedies, (3)
providing funds to assess brownfield sites to promote their economic
redevelopment,2 (4) providing technical assistance so that communities and
tribes located near sites can better participate in cleanup decisions, and
(5) providing for an expanded role for states and tribes in the performance
of the program.

In 1997, we reviewed the 45 reforms from the second and third rounds and
reported that EPA regarded 25 of them, or 56 percent, as fundamental changes
to the Superfund program but could quantify accomplishments for only 6 of
them, or 13 percent.3 EPA stated that, overall, it did not need additional
legislative authority to achieve the reforms' goals but that targeted new
authority would enhance their implementation.

The Government Performance and Results Act of 1993 (the Results Act)
requires that agencies, among other actions, (1) establish strategic plans
containing general goals for the agencies and (2) prepare annual performance
plans that establish goals and measures to assess the results of individual
programs. Therefore, in response to requirements of the Act, EPA established
three general goals for the reforms--faster, fairer, and cost-effective
cleanups--and performance measures for a number of the reforms, all of which
support the agency's strategic and annual goals for the Superfund program
overall. The performance measures for the reforms are intended to
demonstrate progress toward achievement of their goals, which include, among
others, increasing the number of sites where the construction of the cleanup
remedy has been completed and maximizing the participation in cleanups of
the parties responsible for contamination at sites.

Improvements in the Program

EPA claims that as a result of the administrative reforms, the program is
fairer and cleanups are 20 percent faster and cheaper.4 The stakeholders we
contacted also commented that overall, after 20 years, they have a better
working relationship with EPA, the agency is fairer in dealing with
responsible parties, and it is easier to use remedies that are, in their
opinion, more reasonable and cost-effective. But stakeholders also had
questions about the extent to which some of the administrative reforms had
really improved the program. We reviewed EPA' s performance measures for
each reform and found that the agency has more measures in place since our
last review, and for a small number of reforms, the measures demonstrate
results such as cost savings. However, EPA cannot directly link the majority
of its reforms to improvements in the program.

According to EPA reform managers, all 62 reforms are important and have
helped to improve the program, but 42 of them involve activities that (1)
did not have a fundamental impact on the program and (2) could not easily be
measured for any results achieved. EPA reform managers identified the
remaining 20 reforms as activities that have had a fundamental effect on the
program. EPA has established performance measures for 14 of them--an
increase since our prior review, when EPA had measures for 6 of its key
reforms. As table 1 illustrates, EPA's measures for all 14 reforms track the
number of times they were implemented, but measures for only 7 reforms
demonstrate how they have improved the program.

Table 1: Fourteen Fundamental/Measurable Reforms and Their Output and
Outcome Measures

Continued

                                     Measures that count
  Fundamental/measurable   Year of       frequency of       Measures that
          reform           reform       implementation       demonstrate
                                          (output)        results (outcome)
                                    EPA has used this
                                    tool at 9-24 sites
                                    each year.

 Encourage greater use of           The number of
 alternative tools for              settlements that EPA
 resolving liability      1993a     reached peaked in
 disputes                           fiscal year 1997 and
                                    declined slightly
                                    the following year.
                                    Data for fiscal year
                                    1999 were not
                                    available.
                                    EPA maintains that
                                    responsible parties
 Promote "enforcement               have funded about
 first"--getting private            70-84 percent of
 parties to fund most of  1993 a    cleanups since
 the cleanups                       fiscal year 1992,
                                    the year before EPA
                                    announced the
                                    reform.
                                    EPA has achieved
                                    more settlements
                                    after the reform
                                    (33-105 each year)
                                    than before the
 Promote more settlements           reform.
 to provide liability
 protection for de                  The number of
 minimis contributors     1993 a    settlements EPA
 (parties that contribute           achieved peaked in
 small amounts of waste)            fiscal year 1997,
                                    significantly
                                    declined in fiscal
                                    year 1998, and only
                                    slightly increased
                                    in fiscal year 1999.
                                    EPA has signed 16-28
                                    agreements each year  EPA facilitated
                                    since the reform.     the purchase of
 Negotiate agreements to                                  1,500 acres of
 provide liability                  The number of         contaminated
 protection for           1995 a    agreements EPA        property and the
 prospective purchasers             achieved peaked in    redevelopment of
 of contaminated property           fiscal year 1997 and  hundreds of
                                    has been declining    thousands of
                                    over the past 2       adjacent acres.
                                    years.
                                    EPA has made from
                                    20-30 compensation
                                    offers each year,
 Provide compensation for           for a total of
 cleanup costs                      $175.3 million.
 attributable to          1995
 insolvent and defunct              EPA has reached
 parties (orphan shares)            agreement on a total
                                    of 47 of these
                                    offers, for $88
                                    million.
                                    EPA has set up 18-33
                                    accounts each year
                                    since the reform,
                                    making over $570
                                    million available
                                    for site-specific
                                    cleanups.

                                    EPA will measure the
 Encourage the use of               number and amounts
 special accounts for               of disbursements
 site-specific cleanup    1995 a    from these acounts
 costs                              beginning in fiscal
                                    year 2000.

                                    The number of
                                    accounts EPA
                                    established peaked
                                    in fiscal year 1997
                                    and has been
                                    declining over the
                                    past 2 years.
 Revise guidance on
 liability protection
 settlements for de
 micromis parties         1995      EPA has made a total
 (parties that contribute           of 16 settlements.b
 miniscule amounts of
 waste)
                                    EPA began updating
                                    remedies as early as
                                    1983 and has updated
                                    61-85 remedies each
                                    year since fiscal
                                    year 1995, the year   EPA estimates the
                                                          net future cost
 Update cleanup remedy              before it announced   savings from the
 decisions to take                  this reform.          updates conducted
 advantage of new science 1995 a                          during fiscal
 and technology                     The number of         years 1996-99
                                    remedies EPA updated
                                    peaked in fiscal      could total $1.3
                                    year 1997, declined   billion.c
                                    in fiscal year 1998,
                                    and remained at
                                    about that level in
                                    fiscal year 1999.
                                    EPA has completed
 Increase the number of             the construction of
 sites where the                    all remedies at
 construction of all      1993 a    61-88 sites each
 cleanup remedies has               year since fiscal
 been completed                     year 1992, the year
                                    before it announced
                                    this reform.

 Establish the National                                   EPA estimates
 Remedy Review Board to             EPA has reviewed      that its reviews
 review high-cost         1995      9-11 cleanup          have saved a
 proposed remedies                  proposals each year.  total of $70.7
                                                          million to date.
                                    EPA has accomplished
                                    12-27
                                    non-time-critical
                                    removals each year
 Use the Superfund                  since fiscal year     EPA estimates
 Accelerated Cleanup                1992, the year        that it saves, on
 Model (which allows the            before it announced   average, about
 use of shorter-term                this reform.          $2,500 and 11
 cleanup actions, called  1993 a                          months by
 removals, and combined             EPA accomplished a    combining
 site assessment                    total of 442          assessments at a
 activities)                        integrated            site.
                                    assessments and 405
                                    combined assessments
                                    through fiscal year
                                    1999.
                                                          EPA estimates
                                                          that over 1,900
                                                          properties have
                                                          been assessed,
                                                          120 have been
                                                          cleaned up, and
 Fund brownfield                    EPA funded a total    169 have been
 assessment pilot         1995 a    of 305 assessment     redeveloped and
 projects                           grants through        that over 5,800
                                    October 1999.         jobs and about
                                                          $1.9 billion of
                                                          private dollars
                                                          have been
                                                          leveraged at
                                                          sites assessed
                                                          with EPA funds.
                                    EPA has helped to     EPA surveyed
                                    form 3-16 community   members of
                                    advisory groups each  communities near
                                    year.                 7 Superfund sites
 Establish community                                      and determined
 advisory groups          1995 a    The number of groups  that 47 percent
                                    EPA established       believe that EPA
                                    peaked in fiscal      is effectively
                                    year 1997 and has     involving them in
                                    been declining over   the Superfund
                                    the past 2 years.     process.
                                    EPA has awarded 4-37
                                    grants each year
                                    since fiscal year
                                    1988 and has
                                    conducted 7-46
                                    technical outreach
                                    projects each year.   EPA surveyed
                                                          members of
 Promote early and more             The number of grants  communities near
 effective community                awarded since the     7 Superfund sites
 involvement (primarily             reform peaked in      and determined
 through technical        1993      fiscal year 1995,     that 47 percent
 assistance grants and              declined the next     believe that EPA
 outreach projects)                 year, and has         is effectively
                                    remained at about     involving them in
                                    that level.           the Superfund
                                                          process.
                                    The number of
                                    outreach projects
                                    established peaked
                                    in fiscal year 1998
                                    and declined in
                                    fiscal year 1999.

aThis was not a new activity, but EPA reemphasized or revised it as part of
the agency's administrative reform effort.

bAccording to EPA, the fact that so few parties have asked for a settlement
means that such parties are no longer being threatened with lawsuits from
larger parties for a share of the cleanup costs. Therefore, in EPA's view,
the reform is a success.

cWe did not verify the accuracy of EPA's savings estimate; a 1997 industry
study cautioned that these savings may be overstated.

Source: GAO's presentation of information from EPA.

EPA's outcome measures demonstrate positive results for seven reforms. The
measures indicate that two reforms have helped the agency move toward its
goal of more cost-effective cleanups by achieving significant dollar savings
on the types of remedies selected at sites. The measures also demonstrate
that five other reforms have achieved positive results, such as an increase
in the number of brownfield sites assessed (since assessment leads to
cleanup and redevelopment) and feelings of greater participation in cleanup
decisions expressed by some communities that received grants, technical
assistance, or outreach from EPA.

For the seven reforms that do not have outcome measures, it is difficult for
EPA to determine how well they are working, whether they need revision to
become more effective, and whether they are achieving their intended
results--faster, fairer, and cheaper cleanups. For example, it is difficult
for the agency to determine from its performance measures whether using
alternative dispute resolution has led to settlements with responsible
parties that are fairer, take less time, and reduce legal costs. It is also
difficult for the agency to determine, just by counting how many times a
reform has been implemented each year, the extent to which the reform has
become a routine part of the overall program.

EPA reform managers acknowledged that it is very difficult to set
performance measures that directly demonstrate the extent to which the 14
reforms are achieving their goals. The managers pointed out that a number of
reforms, such as those addressing the remedies selected at a site, work
together to cumulatively benefit the program and the agency cannot
separately measure the contribution of each reform. The managers further
acknowledged that factors other than the reforms themselves likely
contributed to the benefits the agency attributes to some of the reforms.
For example, an agencywide policy on the use of alternative dispute
resolution across all EPA programs, not the Superfund reform alone, helped
to increase the use of this technique, and the agency cannot measure the
success of this reform alone.

In a November 1997 internal review of the reforms, EPA acknowledged
limitations in its performance measures and agreed that it needed to do more
than count how many times a reform has been implemented to determine its
results. Furthermore, when the agency has tried to improve its evaluation of
a reform, its efforts have paid off. Specifically, it has learned in some
instances that a reform was not working as well as intended and needed to be
improved. For example, in 1999, EPA completed the first phase of an ongoing
effort to measure the effects of its community involvement reforms. One of
its findings was that only about half of those surveyed considered the
agency effective in involving their communities in the Superfund process,
leading the agency to conclude that it needed to improve its implementation
of these reforms. To its credit, EPA is taking actions to evaluate the
performance of the reforms overall, as well as of certain individual
reforms. The agency is about to update its 1997 internal review of the
reforms to develop a strategy to improve their implementation. In addition,
it is currently compiling the results of a survey it conducted with 36
property buyers to determine how effectively its agreements with these
buyers to limit their liability under Superfund law have helped to stimulate
the reuse of their properties.

The agency recognizes that to fully evaluate the results of reforms, it
needs input from responsible parties. One way of obtaining this input is by
surveying parties on the reforms. However, under the Paperwork Reduction
Act, EPA cannot survey more than nine members of the public without the
approval of the Office of Management and Budget (OMB). In 1999, EPA asked
OMB for general authority to conduct up to 15 separate surveys of
responsible parties' experiences with the reforms. OMB denied the request,
in part because it did not specify how EPA planned to collect and analyze
the data. However, OMB encouraged EPA to resubmit its request after it had
developed a statistical data collection and analysis plan, among other
things. EPA managers said they are trying to decide how to respond to OMB's
suggestions, given the agency's limited resources for contractors to perform
surveys and competing priorities for these resources. During March of this
year, however, EPA did obtain general agencywide authority from OMB to
conduct customer satisfaction surveys. This authority may be sufficient for
the reform managers to survey responsible parties on the reforms.

EPA's data on the 14 reforms also showed two trends indicative of limits on
the progress achieved to date and possibly in the future, namely, that EPA
may not be sustaining its implementation of the reforms and that the regions
may be inconsistent in their use of some reforms. First, EPA's data on the
number of times the 14 reforms have been implemented showed that for almost
half of the reforms, implementation peaked in fiscal year 1997 and then
declined in subsequent years. This suggests that the regions may not be
sustaining the level of implementation achieved after the reforms were
announced and may need additional support or incentives to sustain their
implementation. Alternatively, other factors may be mitigating the effects
of the reforms over time. EPA reform managers acknowledged that the
implementation of some reforms may naturally decline at some time in the
future, when EPA has finished constructing most remedies and is likely to be
bringing fewer sites into the program. We acknowledged in two 1999 reports
that the construction of most remedies at sites currently in the program
would be completed by 2005 and that, because states are now taking on more
of the cleanup workload, fewer sites may come into the program in the
future.5 However, EPA cleanup managers stated that these possible future
trends for the program do not explain the declines in implementation that we
identified for fiscal years 1998 and 1999.

Second, EPA's data also showed that the regions varied widely in the number
of times they used most of the reforms, possibly indicating inconsistency in
their use of the reforms. According to the data, all regions appear to be
giving priority to completing the construction of cleanup remedies. In part,
this is because EPA headquarters has made this a top priority for the
Superfund program, has monitored the regions' implementation closely, and
this past year for the first time made mid-year regional budget adjustments
to reward the regions that were achieving this goal. However, the data for
other reforms do not appear to show consistent levels of implementation
across the regions. For example, regions II and V have relatively large
portions of the overall Superfund workload to manage--17 and 20 percent,
respectively. Yet Region V implemented a significantly higher portion of the
total non-time-critical removals and combined site assessments than Region
II, as figure 1 illustrates. Likewise, regions II and V used special
accounts and removals less frequently or as often as regions VII and VIII,
yet these latter two regions had much smaller portions of the Superfund
workload. Such regional variation could indicate that certain regions are
not realizing the potential savings in time and costs expected from the
reforms.

Figure 1: Comparison of Four Regions' Superfund Workloads and Rates of
Implementation for Three Reforms
Source: GAO's presentation of data from EPA.

EPA reform managers pointed out that factors other than the sizes of the
regions' Superfund workloads may account for the variation and are beyond
the regions' control. For example, for several reforms, such as those to
enhance community involvement in cleanups, the regions can offer the
reforms, but it is up to stakeholders to pursue them. Or, one region may
have fewer sites involving multiple responsible parties, so that region may
have fewer opportunities to use some of the reforms aimed at achieving
faster, cheaper settlements. However, without determining why such
significant variations exist among the regions, EPA cannot be sure that its
reforms are being used to the maximum extent possible.

Furthermore, EPA reform managers acknowledged that ensuring regional
consistency is a constant challenge for the agency and that some regions
were quicker than others to embrace the reforms. In fact, in a 1997 review
of the reforms, EPA itself identified the need to ensure better commitment
to the reforms. The EPA managers noted that differences in the
organizational structures and leadership of the regions could lead to
inconsistencies in implementing the reforms. Likewise, industry and state
cleanup agency officials expressed concerns that some regions, and even
cleanup managers within regions, are less willing than others to implement
certain reforms. These officials felt, therefore, that they could not
realize the full benefits of the reforms, such as lowering litigation and
cleanup costs.

EPA reform managers in headquarters and in the two regions we contacted
outlined EPA's current methods to help ensure that the regions implement the
reforms. These include the use of headquarters liaisons to the regions who
monitor the regions' progress towards annual targets--the number of times
the regions implement certain reforms--and conduct regional visits,
conference calls, and training sessions to discuss the reforms. EPA has also
issued new or updated guidance on the use of some of the reforms. By better
targeting these methods, EPA could more fully achieve the reforms' goals
across the regions.

EPA and some stakeholders we contacted--officials representing industry,
state cleanup agencies, and environmental and community groups--agreed on
the benefits of establishing some of the administrative reforms in law but
disagreed on the need to do so for other reforms. More specifically, the
stakeholders preferred that reforms intended to provide liability protection
to certain parties, such as prospective property purchasers, be established
in law. Stakeholders worried that otherwise, EPA regions had too much
discretion to decide which parties would benefit and affected parties did
not have a firm basis to challenge these decisions. EPA managers explained
that the agency would support legislation, if proposed, to provide liability
relief for such parties but that the agency itself is not currently seeking
any legislation to codify its reforms. According to EPA, it would support
such proposals because they would (1) give such parties greater assurance
that they would not be held liable for the costs of a cleanup under
Superfund, (2) reduce the parties' legal costs, and (3) promote the
development of brownfields, since the fear of being held liable under
current Superfund law can deter parties from pursuing brownfield cleanups
and redevelopment.

Stakeholders and EPA both favored legislation that would provide liability
protection for

� prospective purchasers of contaminated property,

� landowners who were not responsible for or aware of contamination on their
property (innocent landowners),

� owners of property contiguous to a contaminated site, and

� small municipal waste generators and transporters.

Both EPA and the officials representing industry would also like the agency
to be able to compensate parties more extensively for the shares of cleanup
costs attributable to insolvent or defunct parties as a means of promoting
faster and less costly settlements. However, EPA cleanup managers said that
the agency could not afford to do this without obtaining additional funding
authority for this purpose from the Congress. The managers said the agency
continues to request additional funds for the Superfund program that would
allow it to devote more resources to covering such shares of cleanup
costs--$150 million in fiscal year 2001--but have not yet obtained such
funds.

EPA and stakeholders did not agree on the need for other legislative
changes. For example, EPA did not agree with the executive director of the
organization representing small businesses on the need for further
legislative authority to protect such businesses. The agency maintains that
its administrative reforms aimed at removing small waste contributors from
lengthy settlement negotiations, protecting them from litigation, and
adjusting their settlement costs on the basis of their ability to pay
provided these businesses with ample relief. The executive director
acknowledged that these reforms were helpful but said that some member
businesses still report incurring high legal costs that threaten their
financial viability. Therefore, these businesses would like the liability
protection and other benefits of the reforms established in law to make them
less discretionary and further reduce costs.

EPA and stakeholders also differ on how much liability relief should be
extended to parties that conduct cleanups under state programs. In general,
officials representing industry and the states explained that the fear of
being held liable under the current Superfund law deters parties that would
voluntarily clean up sites under state programs, especially brownfield
sites.6 EPA has maintained that it cannot provide parties that clean up a
site under a state program with full relief from Superfund liability. But
the agency can provide these parties with assurances that it no longer has
any interest in the site unless it presents an imminent and substantial
endangerment to public health or the environment in the future. The industry
and state officials believe that this qualified relief from liability is not
sufficient to overcome barriers to cleanups and that a legislative solution
may be necessary.

Several bills that would exempt various parties from liability and therefore
would limit the potential sources of funding for cleanup costs have been
introduced in the Congress in recent years. While some proposed bills to
reauthorize the Superfund program would reinstate the expired Superfund
taxes, others would not. The Congress has not passed any of these bills.

EPA claims and stakeholders agree that the Superfund program is working
better and that, at least collectively, the administrative reforms have
played some part in this improvement, but the agency has not measured the
impact of most reforms. This limits the agency's ability to determine how
well the reforms are working and where it may need to adjust its reform
efforts. EPA's ability to better measure the results of its reforms could be
further limited if the agency does not obtain important data and input from
the responsible parties that are conducting a majority of cleanups, as well
as other key stakeholders, such as community and environmental groups.
Furthermore, without sustaining the most important reforms and ensuring that
all regions are implementing them to the maximum extent possible, the agency
is not assured that it is fully achieving potential benefits, such as saving
significant cleanup dollars and cleaning up sites more quickly. Therefore,
as the agency updates its internal review of the reforms and develops a
reform strategy, it has the opportunity to consider ways that it could
better (1) measure the results of the most important reforms and (2) verify
that it does not have a problem with inconsistent regional implementation
for some reforms.

To achieve the maximum benefits possible from the Superfund administrative
reforms, the Administrator, EPA, should direct the Assistant Administrator
for Solid Waste and Emergency Response, who manages the Superfund program,
to address, in EPA's internal review and update of the reforms, ways in
which the agency can

� cost-effectively obtain additional data--for those reforms with the
greatest potential for improving the program--that would help it better
assess the reforms' results, including continuing to pursue authority from
OMB to solicit input from private parties and other key stakeholders on the
success of the reforms, and

� target incentives or other strategies as necessary to sustain the
implementation of some reforms and better understand whether regional
variation in their use reflects inconsistencies that need to be addressed.

We provided copies of a draft of this report to EPA for its review and
comment. EPA's comments are reproduced in appendix IV. EPA said that it
would evaluate our recommendations and include them in its Superfund reforms
strategy, as appropriate. However, EPA had three principal concerns about
our findings. While we acknowledge the agency's position on these issues, as
discussed individually below, we continue to believe that our findings were
soundly developed and fairly presented. Therefore, we did not change our
report in response to these concerns. Specifically:

� The agency regards all 62 reforms as important and believes that they have
improved the program, even if the precise results of many cannot be
measured. EPA said that it had designated 20 of the 62 reforms as
fundamental because they had the biggest impact, individually, on the
program, but that many of the remaining 42 reforms work together with the
fundamental reforms to improve specific aspects of the program, such as
remedy selection. We had already noted in the report that the agency
considered all reforms to be important and beneficial to the program and
that the agency believed certain reforms worked together to improve the
program, even though the agency could not measure their individual
contributions to the improvements.

� The agency disagreed with our finding that several of the reforms it
designated as fundamental have not produced measurable outcomes. The agency
also stated that it is difficult to measure progress toward certain goals,
such as greater fairness in the program and lower litigation costs, but
there are a number of indicators of this progress. In addition, EPA said
that it has been unable to obtain the authority from OMB to survey private
parties on the reforms' accomplishments. In assessing these accomplishments,
we asked the agency to provide us with any data that it had to demonstrate
results. We took these data and used two criteria to designate whether the
data represented output or outcome performance measures: (1) the standard
definition under the Results Act that an output measure counts activities
undertaken while an outcome measure assesses the results of a program
activity compared to its intended purpose, and (2) the extent to which the
performance measure directly assessed progress toward or achievement of
EPA's stated goals for a reform. Subsequently, we found that our designation
of EPA's performance measures as measuring either activities conducted or
results achieved was consistent with the way the agency itself characterized
them in its issued work plan for Superfund, generated in response to the
Results Act. Furthermore, we had already acknowledged in the report some of
the difficulties the agency faced in measuring progress toward the reforms'
goals and attempting to obtain authority to ask stakeholders for important
data that the agency needed to measure the reforms' results.

� The agency maintains that it has sustained a high level of commitment to
implementing the reforms. Furthermore, the agency stated that the trend data
cited in the report indicating possible declines and regional variation in
the implementation of some reforms over the past several years do not
demonstrate a decrease in the agency's commitment but could reflect the
impact of other factors. These include factors such as annual differences in
the types and number of cleanup activities being conducted in a particular
region, or an overall decline in the cleanup workload as more sites progress
through the cleanup process. These factors could also include ones that the
agency cannot control, such as different levels of interest among
stakeholders in using community advisory groups or technical assistance
grants. We had already acknowledged in the report that the trends showing
variation in implementing the reforms could be due to a number of factors.
One of these factors was not, however, an overall decline in the cleanup
workload. As we point out, the agency itself had admitted that such a
decline could affect reform accomplishments in the future, but does not
explain the decrease in accomplishments over the past several years.
Furthermore, our point is that without good performance data, the agency
cannot know if certain trends indicate implementation problems that the
agency needs to address or are due to factors outside the agency's control.
We showed that when the agency has obtained data from stakeholders on the
reforms' accomplishments, it has learned valuable information about
implementation problems and taken subsequent action to address them.
Therefore, we believe that by focusing on the most critical reforms and
significant variation in their implementation and verifying the root cause
of this variation, the agency could achieve similar improvements in these
reforms.

In addition to these overall comments, EPA provided technical and clarifying
comments that we incorporated in the report as appropriate.

Unless you announce its contents earlier, we plan no further distribution of
this report until 30 days after the date of this letter. At that time, we
will send copies of the report to appropriate congressional committees and
interested members of the Congress. We will also send copies of this report
to the Honorable Carol M. Browner, Administrator, EPA; and the Honorable
Jacob Lew, Director, Office of Management and Budget, and we will make
copies available to others on request. Please contact me at (202) 512-6111
if you or your staff have any questions. Key contributors to this report are
listed in appendix V.
Peter F. Guerrero
Director, Environmental Protection
Issues

List of Requesters

The Honorable Thomas J. Bliley, Jr.
Chairman, Committee on Commerce
House of Representatives

The Honorable Michael G. Oxley
Chairman, Subcommittee on Finance
and Hazardous Materials
Committee on Commerce
House of Representatives

The Honorable Sherwood L. Boehlert
Chairman, Subcommittee on Water
Resources and Environment
Committee on Transportation
and Infrastructure
House of Representatives

Fourteen Reforms That EPA Considers as Fundamental and as Having Achieved
Measurable Results

This appendix summarizes our analysis of the 14 administrative reforms that
the Environmental Protection Agency (EPA) characterized as having
significantly and measurably improved the Superfund program. For each of the
reforms, our analysis considers the performance goals and measures and the
results identified by EPA, our own and stakeholders' observations, and EPA's
and others' views on the need for any additional authority to implement the
reform.

In 1993, EPA initiated a reform to encourage parties at Superfund sites to
use alternative methods for resolving disputes over liability for cleanup
costs. These methods--used by the agency since 1987--include employing
neutral parties to organize negotiations to allocate cleanup costs among all
involved parties and facilitate settlement deliberations. EPA initiated
training for staff and allocated funds to its regions for alternative
dispute resolution programs. Table 2 shows EPA's performance goals and
measures for this reform.

Table 2: Performance Goals and Measures for Encouraging the Use of
Alternative Dispute Resolution

                                           EPA's performance measures
 Reform                  Goals             Outcomes  Outputs
                                                     Number of Superfund
                                                     enforcement
                         Increase                    cases/sites using
 Encourage greater use   program's                   alternative dispute
 of alternative dispute  fairness                    resolution: 133 since
 resolution in settling                    None      1987
 disagreements over      Reduce litigation
 Superfund liability     and associated              Percentage of cases
                         costs                       using alternative
                                                     dispute resolution
                                                     that reach settlement:
                                                     78-80 percent

Source: GAO's classification and presentation of information from EPA.

Figure 2 shows how many Superfund liability cases EPA initiated each year
using alternative dispute resolution.

Figure 2: Superfund Liability Cases Initiated Using Alternative Dispute
Resolution, Fiscal Years 1993-98
Note: Fiscal year 1998 was the last year for which EPA was able to provide
complete data on this reform.

Source: GAO's presentation of data from EPA.

EPA's data for fiscal years 1993-98 indicate that all regions have used
alternative dispute resolution techniques in resolving Superfund disputes
(see fig. 3).

Figure 3: Each Region's Percentage of National Superfund Cases in Which
Alternative Dispute Resolution Was Used, Fiscal Years 1993-98
Source: GAO's presentation of data from EPA.

� Sustained effects: The data show that the use of alternative dispute
resolution techniques has increased since the reform was initiated. However,
the use of these techniques decreased in fiscal year 1998. The data are
insufficient to determine if this is a long-term trend.

� Contribution of other factors: EPA reform managers acknowledge that other
initiatives--including an agencywide policy and several laws7 authorizing
the use of alternative dispute resolution techniques--have contributed to
the greater use of these techniques in Superfund cases, but they cannot
isolate the effects of either the reform or these other factors.

� Regional implementation: The data show variations among the regions in
implementing this reform that do not correspond with differences in the
sizes of their Superfund workloads. According to EPA reform managers, one
possible explanation for these variations is that not all sites are
appropriate for the use of alternative dispute resolution techniques.
However, the agency cannot determine the number of sites that might be
appropriate for using the techniques.

� Performance measurement:

� Counting the number of times alternative dispute resolution is used does
not directly demonstrate that the program is fairer and that litigation and
its associated costs have decreased. Nevertheless, the increased use of
alternative dispute resolution could imply that the program is "fairer" than
it would have been otherwise because all parties are satisfied with how
cleanup costs are allocated under the agreements reached.

� EPA does not (1) track litigation costs; (2) have access to this
information; and (3) according to EPA reform managers, have authority to
require private parties to provide it.

EPA reform managers do not believe that the agency needs any additional
legal authority to pursue this reform.

In 1990, EPA began to emphasize more vigorous Superfund enforcement to
increase the proportion of cleanup actions funded by responsible parties.
EPA included "enforcement first" among the administrative improvements it
announced in fiscal year 1993. Under this initiative, EPA seeks commitments
from responsible parties to fund and perform at least 70 percent of all new
remedial action work at Superfund sites. Table 3 shows EPA's performance
goals and measures for this reform.

Table 3: Performance Goals and Measures for Promoting "Enforcement First"

                                            EPA's performance measures
 Reform               Goals                 Outcomes  Outputs
                                                      Percentage of total
                      Increase program's              cleanup actions
                      fairness                        funded by responsible
 Emphasize                                            parties: About 70
 enforcement to       Expedite cleanups               percent as of fiscal
 encourage cleanups                         None      year 1999 (GPRAa
 by responsible       Increase the                    measure)
 parties              proportion of
                      cleanups funded by              Amount committed by
                      responsible parties             parties toward
                                                      cleanups since 1980:
                                                      $16.2 billion

aAccording to EPA, these measures track and/or project the number of actions
that occur throughout the year (accomplishments) and are used to evaluate
the program's progress in support of the Government Performance and Results
Act of 1993 (GPRA or the Results Act).

Source: GAO's classification and presentation of information from EPA.

Figure 4 shows the percentage of new cleanup actions funded by responsible
parties during fiscal years 1987-99.

Figure 4: Percentage of New Cleanup Actions Funded by Responsible Parties,
Fiscal Years 1987-99
Source: GAO's presentation of data from EPA.

Figure 5 shows, for each EPA region, the average percentage of new cleanups
started by responsible parties during fiscal years 1995-99, compared with
the national average of 70 percent.

Figure 5: Average Percentage of New Cleanups Funded by Responsible Parties,
by Region, Compared With the National Average, Fiscal Years 1995-99
Source: GAO's presentation of data from EPA.

� Sustained effects: EPA has, over time, increased the proportion of new
cleanup actions financed by responsible parties and, in recent years,
maintained this proportion at about 70 percent or higher. Both EPA reform
managers and industry representatives credit the reform with these results.

� Contribution of other factors: EPA reform managers identified no other
factors as contributing to the results credited to this reform.

� Regional implementation: Regional data indicate that the regions vary in
the extent to which they are obtaining responsible parties' commitments to
fund new cleanup activities initiated each year. For each region,
responsible parties' rate of participation in cleanups varies from year to
year, ranging from 0 to 100 percent. These variations do not correspond with
differences in the sizes of the regions' Superfund workloads.

� Performance measurement: Measuring the rate at which responsible parties
pay for cleanups does not directly measure progress toward or achievement of
this reform's goal of increasing the program's fairness. Nevertheless,
ensuring that responsible parties are now paying for a larger portion of the
Superfund cleanup work than they did earlier in the program could imply that
the program is "fairer" than it would have been without the reform.

EPA reform managers do not believe that the agency needs any additional
legal authority to pursue this reform.

In 1993, EPA began to emphasize efforts to resolve the liability of de
minimis parties as early as possible in the Superfund process. According to
EPA, de minimis parties are those whose contributions of hazardous waste to
a Superfund site are minimal in volume and toxicity. With this reform, EPA
simplified its requirements for determining parties' eligibility for such
settlements, streamlined the settlement process, and issued guidance to its
regions encouraging these settlements. Table 4 shows EPA's performance goals
and measures for this reform.

Table 4: Performance Goals and Measures for Promoting Settlements With Small
Waste Contributors

                                           EPA's performance measures
 Reform                Goals               Outcomes  Outputs

                       Reduce litigation             Number of settlements
 Encourage more        and associated                with de minimis
 liability settlements costs                         parties: 465 as of the
 with parties that have                              end of fiscal year
 contributed small     Increase program's            1999 (GPRA measure)
 amounts of waste at a fairness            None
 site (de minimis                                    Number of parties
 parties)              Encourage more,               released from the
                       early, and                    Superfund liability
                       expedited                     system: Over 21,000 as
                       settlements                   of the end of fiscal
                                                     year 1999

Source: GAO's classification and presentation of information from EPA.

Figure 6 shows the number of de minimis settlements per year during fiscal
years 1987-99.

Figure 6: Number of De Minimis Settlements, Fiscal Years 1987-99
Source: GAO's presentation of data from EPA.

EPA's data show that all regions have completed de minimis settlements (see
fig. 7).

Figure 7: Each Region's Percentage of National De Minimis Settlements,
Fiscal Years 1987-99
Source: GAO's presentation of data from EPA.

� Sustained effects:

� Although the number of de minimis settlements has generally increased
above prereform levels, this number has fluctuated significantly from year
to year since the reform began. The number of settlements EPA achieved
peaked in fiscal year 1997 and subsequently declined. Although the data are
insufficient to determine if this is a long-term trend, these fluctuations
could suggest that EPA may not be sustaining the implementation of this
reform.

� EPA reform managers stated that the number of de minimis settlements for
fiscal year 1997 is unusually high because (1) 42 of these settlements
involved parties at a single site who each wanted an individual settlement
and (2) 25 settlements involved de minimis landowners at another site who
were provided protection for $0. According to the reform managers, these two
site-specific experiences, which appear to have dramatically increased the
number of settlements in fiscal year 1997, were anomalies. When the de
minimis settlements for these two sites are removed, 40 such settlements
remain for the year--a total that is more in line with EPA's national
average and GPRA target, according to these managers.

� Contribution of other factors: EPA reform managers identified no other
factors as contributing to the results credited to this reform.

� Regional implementation: The data show variations among the regions in
implementing this reform that do not correspond with differences in the
sizes of their Superfund workloads. According to EPA reform managers, one
possible explanation for these variations is that de minimis parties are not
involved at all sites, and the mix of sites with such parties may vary from
region to region and year to year.

� Performance measurement: EPA's measures of this reform's success do not
directly indicate progress toward or achievement of its goals--increasing
the program's fairness and reducing litigation and its associated costs.
However, because de minimis settlements remove small or innocent parties
from the liability allocation process and shield them from costly
litigation, the settlements could imply that the program is "fairer" than it
would have been otherwise.

� Views on the reform's benefits/effectiveness:

� The executive of the organization representing independent businesses
stated that, in his opinion, based on information from member companies,
EPA's de minimis settlements do not sufficiently protect landowners from
Superfund cleanup liability because these parties incur high costs to reach
de minimis settlements. However, EPA disagrees, maintaining that the
majority of de minimis parties have paid less than $5,000 each to resolve
their liability.

� Industry representatives felt that removing de minimis parties from the
liability allocation process early would shield them from paying their fair
share of cleanup costs because they would not incur responsibility for any
cost overruns that might occur later in the cleanup. EPA reform managers
maintain that de minimis parties pay a premium to settle early and that this
premium is often in excess of any cost overruns that occur.

� Some small businesses would like to see de minimis protection established
in law, including a provision that would treat a party's ability to pay as a
mitigating factor in the settlement, according to the executive of the
organization representing independent businesses.

� EPA disagrees that additional legislation is needed and argues that it has
sufficient authority, models, and guidance in place to successfully
accomplish de minimis and ability-to-pay settlements.

In May 1995, EPA revised its guidance on agreements with prospective
purchasers. In such an agreement, EPA promises not to sue a purchaser for
contamination that the purchaser did not cause in exchange for the
purchaser's commitment to perform cleanup work or provide funds toward
cleaning up the site. EPA (1) revised the criteria for evaluating whether it
should negotiate an agreement, (2) broadened the universe of sites at which
it would consider negotiating an agreement, and (3) encouraged a more
balanced trade-off between the benefits to EPA and the public from these
agreements. Table 5 shows EPA's performance goals and measures for this
reform.

Table 5: Performance Goals and Measures for Reaching Agreements With
Prospective Purchasers

                                     EPA's performance measures
 Reform             Goals            Outcomes             Outputs
                                                          Number of
                                                          prospective
                                     Facilitated the      purchaser
                                     purchase of over     agreements
 Provide assurances Increase         1,500 acres of       signed: 114 since
 to prospective     program's        contaminated         1989
 purchasers of      fairness         property
 contaminated                                             Number of
 property that they Restore          Spurred the          prospective
 will not be held   contaminated     redevelopment of     purchaser
 liable for cleanup sites to         hundreds of          agreement
 costs              beneficial use   thousands of         requests
                                     adjacent acres       received/
                                     nationwide           addressed (GPRA
                                                          measure beginning
                                                          in fiscal year
                                                          2000)

Source: GAO's classification and presentation of information from EPA.

Figure 8 shows the number of prospective purchaser agreements completed
since fiscal year 1989.

Figure 8: Number of Prospective Purchaser Agreements Completed, Fiscal Years
1989-99
Source: GAO's presentation of data from EPA.

Regional data indicate that all regions are implementing this reform,
although at varying rates, as shown in figure 9.

Figure 9: Each Region's Percentage of National Prospective Purchaser
Agreements, Fiscal Years 1989-99
Source: GAO's presentation of data from EPA.

� Sustained effects: The number of prospective purchaser agreements has
increased since the reform was initiated. However, the number of agreements
peaked in fiscal year 1997 and has declined during the past 2 years. This
decline may indicate that EPA is not sustaining its implementation of this
reform. It is difficult for EPA to determine from its performance measures
whether the reform has had a significant effect on efforts to redevelop
brownfield properties.

� Contribution of other factors: EPA reform managers acknowledge that
factors other than this reform--particularly the agency's focus on the reuse
of contaminated properties-- contributed to the overall increase in the
number of agreements.

� Regional implementation: The data show variations among the regions in
implementing this reform that do not correspond with differences in the
sizes of their Superfund workloads. According to EPA reform managers, one
possible explanation for these variations is that the number of agreements
is determined by responsible parties' requests for such agreements, over
which EPA has no control. Consequently, according to these managers, the
number of agreements for each region would not be expected to be
proportional to the size of its workload.

� Performance measurement:

� An increase in the number of prospective purchaser agreements completed
does not directly demonstrate that EPA is achieving this reform's
goals--increasing the program's fairness and restoring sites to beneficial
use. However, because these agreements absolve purchasers of contaminated
property who did not cause the contamination of future cleanup liability,
the agreements could imply that the program is "fairer" than it would have
been otherwise. Similarly, the number of agreements could imply that this
reform contributes to property redevelopment because, by limiting liability,
an agreement makes the purchase of a contaminated property more economically
attractive, encouraging potential buyers to purchase and redevelop it.

� In fiscal year 2000, EPA began tracking the number of agreement requests
received and assessed as a measure of the reform's performance. These data
will not directly measure the reform's effect on the program.

� EPA recognized that it needed to measure the reform's effectiveness more
directly. Accordingly, in the summer of 1998, EPA surveyed the regional
personnel and private parties who had negotiated the 85 prospective
purchaser agreements in existence at that time. Thirty-six of the 85 private
party representatives (42 percent) participated in the survey. The findings
are being compiled for a summary report, to be available in 2000.

� Representatives of the responsible parties we contacted believe that this
reform does not provide the permanent assurances against liability that they
need to make purchases of contaminated property economically attractive.
They maintain that legislation is needed to provide full protection.

� EPA supports legislation that would provide relief from liability to
prospective purchasers of contaminated property, as well as innocent
landowners, contiguous property owners, and small municipal waste generators
and transporters.

Defunct Parties (Orphan Shares)

In October 1995, EPA began compensating parties who agree to perform cleanup
activities at a Superfund site for some or all of the cleanup costs
attributable to other insolvent or defunct parties. Because these parties
cannot pay or are no longer available to pay the costs for which they are
responsible, their shares of the cleanup costs are known as the orphan
shares. EPA developed interim guidance on compensation for orphan shares,
limiting it to (1) 25 percent of the cleanup remedy or removal costs, (2)
the total past and future oversight costs, or (3) the amount of the actual
orphan shares, whichever is less. Table 6 shows EPA's performance goals and
measures for this reform.

Table 6: Performance Goals and Measures for Providing Compensation for
Orphan Shares

                                         EPA's performance measures
 Reform                 Goals            Outcomes Outputs
                                                  Number of compensation
                                                  offers/ dollar amounts
                                                  offered: 98 offers/$175.3
                                                  million offered through
                                                  fiscal year 1999 (GPRA
                                                  measure)
 Compensate responsible Increase
 parties who agree to   program's                 Number of compensation
 clean up a site for    fairness                  agreements reached/dollar
 costs owed by                           None     amount of compensation
 insolvent parties      Reduce                    provided: 47
 (orphan shares)        litigation and            agreements/$88 million
                        associated costs
                                                  through the first quarter
                                                  of fiscal year 2000

                                                  Maximum amount
                                                  appropriate for
                                                  compensation

Source: GAO's classification and presentation of information from EPA.

Figure 10 shows the number of orphan share compensation offers made through
fiscal year 1999.

Figure 10: Number of Orphan Share Compensation Offers, Fiscal Years 1996-99
Source: GAO's presentation of data from EPA.

Figure 11 shows the percentage of orphan share compensation offers made by
each EPA region through fiscal year 1999.

Figure 11: Each Region's Percentage of National Orphan Share Compensation
Offers, Fiscal Years 1996-99
Source: GAO's presentation of data from EPA.

� Sustained effects: The relatively steady trend in the number of orphan
share offers made over the period reflects the number of sites eligible for
the reform. The amount of orphan share compensation relates to the amount of
past costs and future oversight costs available to compensate.

� Contribution of other factors: EPA reform managers identified no other
factors as contributing to the results credited to this reform.

� Regional implementation:

� The limited funds EPA has had each year to offset the costs it assumes
during settlements make it difficult to determine if the regions could be
more extensively implementing the reform.

� EPA reform managers suggested that differences in the regions'
implementation of this reform do not necessarily indicate inconsistent
implementation because the types of sites and costs of cleanup vary across
the country. For example, regions VIII, IX, and X made fewer offers because
the sites are owner- and/or operator-only sites that EPA excludes from
compensation.

� Performance measurement:

� The number of offers and the dollar amounts offered do not directly
demonstrate that EPA has achieved or made progress toward the reform's
goals--increasing the program's fairness and reducing litigation and its
associated costs. Nevertheless, offering to compensate parties for costs
that are attributable to parties who contributed to a site's contamination
but cannot pay for the cleanup could imply that the program is "fairer" than
it would have been otherwise.

� Data on the number of settlements reached and the compensation paid would,
at best, indirectly measure the reform's results.

� Views on reform's benefits/effectiveness:

� Industry representatives believe this reform can lead to faster
settlements. However, they believe that the limits EPA has set on orphan
share compensation can discourage settlements. Some representatives believe
that (1) they are not any better off under the reform; (2) EPA is less
likely to compensate large, financially viable responsible parties; (3) EPA
is inconsistent in deciding what portion of cleanup costs are orphan shares;
and (4) EPA designates some parties as viable, even though the site itself
is their only asset.

� Some parties would like additional compensation and would like EPA to be
more forthcoming about factors such as dollar constraints that limit its
compensation decisions.

� EPA reform managers said that the agency cannot afford to provide
additional compensation without obtaining more funding from the Congress.
Moreover, according to the managers, parties are not worse off under EPA's
policy because the agency can still forgive past costs over and above the
orphan share limits when negotiating settlements. Finally, the managers
noted, EPA provides orphan share calculations to parties at the beginning of
settlement negotiations so that they know the basis for its decisions. The
managers acknowledged that these calculations are rough but said that
providing more detailed ones would be too costly.

To give responsible parties complete relief from liability for costs that
are not attributable to their activities at a site, EPA reform managers
said, EPA needs a congressional allocation to cover the costs attributable
to the activities of insolvent or defunct parties. In each of its annual
budget requests from fiscal year 1996 through fiscal year 2000, EPA
requested $200 million for orphan share compensation, which it wanted the
Congress to provide in a separate account so that the funding for such
compensation would not affect the funding for cleanup. However, EPA and the
Congress have not been able to agree on a legislative proposal to authorize
additional compensation. In its fiscal year 2001 budget request, EPA
decreased its request for orphan share compensation to $150 million in order
to request funds for other purposes.

In October 1995, EPA began encouraging the greater use of "special
accounts." These can be established with any funds received in a Superfund
settlement for a site and can then be used to conduct or finance cleanup
actions or to reimburse responsible parties for future cleanup actions at
the site. The accounts create incentives for the parties to perform cleanup
work under settlements with EPA. Table 7 shows EPA's performance goals and
measures for this reform.

Table 7: Performance Goals and Measures for Encouraging the Use of
Site-Specific Accounts

                                          EPA's performance measures
 Reform                  Goals            Outcomes  Outputs

                         Increase                   Number of accounts and
                         program's                  amounts of funds
 Promote the greater use fairness                   available: 133
 of site-specific                                   accounts/
 accounts that hold      Encourage                  $570 million available
 funds obtained through  responsible                for site-specific
 settlements with        parties to       None      cleanups
 parties at a site for   settle
 cleanup actions at that                            Number of settlements
 site                    Reduce                     disbursing funds from
                         litigation and             accounts and amounts
                         associated costs           disbursed (beginning in
                                                    fiscal year 2000)

Source: GAO's classification and presentation of information from EPA.

Figure 12 shows the number of special accounts established between fiscal
year 1990 and fiscal year 1999.

Figure 12: Number of Special Accounts Established, Fiscal Years 1990-99
Source: GAO's presentation of data from EPA.

EPA's data show that all regions are implementing this reform (see fig. 13).

Figure 13: Each Region's Percentage of National Special Accounts
Established, Fiscal Years 1990-99, Compared With Its Percentage of National
Superfund Sites
Note: National Superfund sites include final and deleted National Priorities
List sites as of Sept. 30, 1999.

Source: GAO's presentation of data from EPA.

� Sustained effects:

� More new special accounts have been established since EPA's reform than
were established before the reform.

� However, the number of new accounts began increasing a few years before
the reform; therefore, the data do not directly link the increases to the
reform.

� The number of new accounts has not increased continuously: in fiscal year
1998, it decreased by about 30 percent from the previous year and continued
to decline in fiscal year 1999. This drop in the number of accounts may
indicate that EPA is not sustaining its implementation of this reform.

� Contribution of other factors: EPA was unable to identify any factors
other than the reform that could have contributed to either the initial
increase or the subsequent decrease in the number of special accounts.

� Regional implementation: The number of accounts established by some
regions was not proportional to the sizes of their Superfund workloads,
suggesting possible inconsistency in regional implementation (see fig. 13).
Furthermore, it is difficult for EPA to determine from its performance
measures whether the number of accounts--133--relative to the hundreds of
eligible Superfund sites nationwide is significant.

� Performance measurement:

� The numbers of settlements/accounts and the amounts of funds deposited do
not directly demonstrate that EPA is achieving the reform's
goals--increasing the program's fairness and encouraging parties to perform
cleanups. There is no clear correlation between the establishment of
accounts and either the program's fairness or the number of settlements.

� Similarly, the numbers of settlements designating disbursements and the
amounts disbursed do not directly measure either fairness or the reform's
effects on settlements.

� EPA reform managers told us that measuring the effects of funds in special
accounts on responsible parties' decisions to settle and perform cleanups
would be another performance measure to track the reform's outcomes.
However, this information would be difficult for EPA to obtain because the
agency cannot survey such parties without receiving approval from the Office
of Management and Budget (OMB) to collect the information. OMB denied EPA's
initial request, and EPA has not decided whether it will submit a revised
request.

� Views on reform's benefits/effectiveness:

� According to a 1999 audit of the special accounts by EPA's Office of the
Inspector General, (1) EPA's process for reconciling transactions and
balances did not ensure that errors could be identified and corrected and
(2) regional personnel were not always sufficiently aware of the existence
and intended use of these accounts. EPA has agreed to take actions such as
revising guidance and providing the regions with updated information and
instructions on technical aspects of managing the accounts to correct many
of the identified problems.

� Industry stakeholders we contacted said they do not consider this a
significant reform because EPA has not made account funds widely available
to responsible parties. EPA reform managers responded that the original
reform did not include disbursements from the special accounts to
potentially responsible parties. EPA first published guidance on
disbursements to potentially responsible parties in November 1998.

EPA reform managers said no additional legal authority is needed to pursue
this reform.

In June 1996, EPA revised its existing guidance to further prevent de
micromis parties--those who contributed miniscule amounts of waste to a
Superfund site--from incurring legal and other costs that may exceed their
share of the site's cleanup costs. The revised guidance (1) expanded the
number of parties eligible for de micromis settlements, (2) offered no-cost
protective agreements that remove de micromis parties from the liability
process, (3) streamlined and simplified the settlement process, and (4)
clarified and emphasized EPA's intent to protect such parties. Table 8 shows
EPA's performance goals and measures for this reform.

Table 8: Performance Goals and Measures for Revised Guidance on Settlements
With Miniscule Waste Contributors

                                           EPA's performance measures
 Reform                  Goals             Outcomes  Outputs

 Revise guidance to      Increase
 discourage litigation   program's                   Number of settlements
 against de micromis     fairness                    with de micromis
 parties--those who                        None      parties: 16 at 11
 contributed miniscule   Reduce                      sites through fiscal
 waste volumes to sites  litigation and              year 1999
                         associated costs

Source: GAO's classification and presentation of information from EPA.

During fiscal years 1993-98, EPA completed 16 de micromis settlements with
parties at 11 Superfund sites. The settlements at five of the sites were
based on EPA's 1993 de micromis guidance, drafted before the reform, and
those at the remaining six sites were based on the agency's 1996 guidance,
revised after the reform. Region III had the largest number of sites with de
micromis settlements (five) through fiscal year 1999; Region II had three
settlements; Regions I, IV, and IX had no settlements; and the remaining
regions had either one or two settlements each.

� Sustained effects: Because the number of de micromis settlements is small,
it is difficult to determine the extent to which the reform has had an
effect or if these effects have been sustained.

� Contribution of other factors: EPA reform managers identified no other
factors as contributing to the results credited to this reform.

� Regional implementation: Because so few settlements have been completed to
date, the data are insufficient to determine if the regions are implementing
the reform consistently.

� Performance measurement: EPA's measures of the success of this reform do
not directly indicate the extent to which the reform is achieving its
goals--increasing the program's fairness and reducing litigation and its
associated costs. However, because de micromis settlements remove parties
that have contributed only miniscule amounts of waste at a site from the
liability process, thereby shielding them from costly litigation, these
settlements could imply that the program is fairer than it would have been
otherwise and that the potential for litigation and any associated costs has
been reduced.

� Views on reform's benefits/effectiveness:

� EPA reform managers believe that the reform is successful if EPA
negotiates only a few de micromis settlements. They believe that the reform
deters responsible parties from filing third-party contribution suits to get
the small parties to help pay for cleanups. If the reform has such a
deterrent effect, small parties do not need a de micromis settlement for
protection.

� Industry representatives agreed that EPA is now more successful in
removing de micromis parties from the liability process.

� The executive of the organization representing small businesses believes
that some miniscule contributors may continue to incur often-onerous legal
costs. EPA reform managers acknowledge that such contributors may incur
legal expenses if either EPA does not identify them as de micromis parties
or they do not identify themselves to EPA in a timely manner.

� EPA reform managers said the agency does not need any additional legal
authority to pursue this reform.

� Some private parties we contacted believe that de micromis liability
protection needs to be established in law to shield miniscule contributors
from onerous legal costs.

In October 1995, EPA began encouraging the regions to revisit cleanup
remedies selected in the past and, where appropriate, to select different
remedies that incorporate recent technological advances, ensuring that the
updated remedies are both protective and more cost-effective. EPA issued
guidance on updating remedy decisions in September 1996. Table 9 shows EPA's
performance goals and measures for this reform.

Table 9: Performance Goals and Measures for Updating Remedy Decisions

                                            EPA's performance measures
 Reform             Goals                   Outcomes        Outputs

 Encourage the                              Total
 regions to revisit                         estimated
 previous remedy    Enhance remedies'       future cost     Total number of
 decisions and use  cost-effectiveness      savings and     remedy updates
 new scientific                             increases for   approved: 295
 information or     Implement remedies that all remedy      during fiscal
 technological      reflect advances in     updates: $1.3   years 1996-99
 advances to update science or technology   billion
 the decisions                              through fiscal
                                            year 1999

Source: GAO's classification and presentation of information from EPA.

Figure 14 shows the number of cleanup remedies updated annually during
fiscal years 1982-99.

Figure 14: Number of Cleanup Remedy Updates per Year, Fiscal Years 1982-99
Source: GAO's presentation of data from EPA.

While EPA reported that the 295 remedies it had updated since the reform
started would result in future cost savings of more than $1.3 billion, the
agency also estimated that some of these updates would result in cost
increases totaling approximately $123 million.

Data from EPA show that all 10 regions are implementing this reform, as
illustrated in figure 15.

Figure 15: Each Region's Percentage of National Cleanup Remedy Updates,
Fiscal Years 1996-99, Compared With Its Percentage of National Superfund
Sites
Note: National Superfund sites include final and deleted National Priorities
List sites as of Sept. 30, 1999.

Source: GAO's presentation of data from EPA.

� Sustained effects:

� EPA's data show that the regions are updating remedy decisions and
achieving cost savings for both responsible parties and EPA. However, the
upward trend in the number of remedy updates during the 7 years preceding
the reform raises questions about the extent to which the reform itself
promoted updating.

� The number of remedies EPA updated peaked in fiscal year 1997, declined in
fiscal year 1998, and remained at about the same level in fiscal year 1999.
The decline in the number of remedy updates since fiscal year 1997 may
indicate that EPA is not sustaining its implementation of this reform.

� EPA predicts that at some point in the future, the regions will have
reviewed most past remedies for an update and the opportunities to achieve
additional cost savings will decrease. But for fiscal year 2000, the regions
plan to review about as many remedies as they have reviewed in the past few
years.

� Contribution of other factors: EPA acknowledges that other reforms
addressing the selection of cleanup remedies, such as a directive providing
more flexibility to assume that sites will be used for industrial rather
than residential purposes and will therefore require less extensive cleanup,
also contributed to the cost savings achieved.

� Regional implementation:

� Some industry representatives said that individual cleanup managers and
regions differ in their willingness to consider a remedy for an update; as a
result, the representatives believe they do not have a consistent chance to
achieve cost savings.

� EPA does not know whether the regions are considering all possible
remedies. EPA's data show regional differences in the number of updates--one
region updated three times as many remedies as another, even though both
regions managed about the same number of Superfund sites.

� EPA did not determine the reasons for the differences. However, according
to EPA reform managers, some regions could have more sites with contaminated
groundwater than other regions. Such sites would be good candidates for
remedy updates because new technologies have become available for cleaning
up contaminated groundwater. According to these reform managers, a remedy
update depends on the availability of new data suggesting that a
modification to the remedy may be feasible; without new data, there is no
basis to change the remedy.

� Performance measurement:

� Measuring the extent of the reform's effect is difficult. EPA admits that
its estimates of cost savings are not rigorous, partly because the agency
depends on private parties to voluntarily provide estimates of cost savings
for the cleanups they manage. The agency believes that it is not
cost-effective to obtain and track more precise cost data to better measure
savings.

� We could not verify the cost savings claimed because EPA (1) does not
maintain supporting documentation for the savings estimates and (2) cannot
require private parties to provide such documentation.

� Views on reform's benefits/effectiveness: Industry representatives gave
high marks to this reform for leading to more cost-effective cleanups.

EPA reform managers said that no additional legal authority is needed to
pursue this reform.

In 1991, EPA began to classify certain sites as construction complete. A
site is considered construction complete when one or more of the following
conditions is met:

� Any necessary physical construction is complete, whether or not final
cleanup levels have been met.

� EPA has determined that the response action should be limited to measures
that do not involve construction.

� A site qualifies for deletion from the National Priorities List (when no
further action is needed to protect human health and the environment).

One of the reforms announced by EPA in 1993 aimed to increase the number of
sites designated as construction complete. Reflecting this priority, in the
mid-1990s, EPA shifted funds from assessing sites whose inclusion in the
Superfund program was not yet certain to completing the construction of
remedies at sites already in the program. Table 10 shows EPA's performance
goals and measures for this reform.

Table 10: Performance Goals and Measures for Increasing Construction
Completions

                                           EPA's performance measures
 Reform                Goals               Outcomes  Outputs
                                                     Number of sites where
 Increase the number                                 all remedies have been
 of sites where all    Accelerate                    constructed: 670
 remedies have been    construction and    Nonea     through the end of
 constructed           advance the program           fiscal year 1999 (GPRA
                                                     measure)

aThe EPA manager for this reform considers the number of sites where
construction has been completed to be an outcome measure. We disagree;
therefore, we show the measure as an output.

Source: GAO's classification and presentation of information from EPA.

Figure 16 shows the number of sites whose cleanup remedies were complete as
of the end of fiscal year 1999.

Figure 16: Number of Sites Designated as Construction Complete, Fiscal Years
1981-99
Source: GAO's presentation of data from EPA.

Each EPA region has completed the construction of cleanup remedies at a
similar rate relative to its share of the nation's Superfund sites (see fig.
17).

Figure 17: Each Region's Percentage of National Construction Complete Sites,
Fiscal Years 1981-99, Compared With Its Percentage of National Superfund
Sites
Note: National Superfund sites include final and deleted National Priorities
List sites as of Sept. 30, 1999.

Source: GAO's presentation of data from EPA.

� Sustained effects: EPA has increased the number of sites designated as
construction complete, in part because of its reform.

� Contribution of other factors:

� The EPA manager for this reform said that other reforms, such as those
promoting the use of more cost-effective remedies or faster settlements with
responsible parties, have also contributed to increases in construction
completions, but the effects of individual reforms cannot be isolated and
measured.

� The maturing of the program, which is now about 20 years old, also
contributed to the increase. With the construction of many remedies taking
more than 10 years to complete, the number of completions was expected to
grow by the early 1990s. Moreover, when EPA shifted resources to this
reform, it also increased the backlog of sites awaiting assessment and
consideration for inclusion in the Superfund program.

� Regional implementation: EPA's data suggest that all regions are
consistently implementing this reform. The number of sites where EPA can
complete construction is limited by the funds available each year for this
purpose.

� Performance measurement:

� EPA measures the increase in construction completions by tracking the
number of sites completed each year. This measure does not necessarily
demonstrate that the agency is completing construction faster.

� In December 1997, EPA's Office of the Inspector General reported that
while EPA had generally reported the construction completion statistic
accurately, it had at times represented sites where construction was
complete as if (1) all cleanup work was done and (2) the sites could be
returned to economic use. However, cleanup work is not always done and sites
cannot always be returned to economic use when construction is complete. As
a result, the Inspector General concluded that EPA might have been
misinforming the Congress and the public as to the status of cleanup at
Superfund sites. In response to a recommendation by the Inspector General,
EPA's Acting Assistant Administrator for Solid Waste and Emergency Response
issued a memorandum emphasizing the need to use the construction completion
statistic accurately in all documents and not to overstate what it
represents.

� Some stakeholders argue that the measure is not meaningful because it
measures only whether remedies have been constructed, not whether the
cleanup itself is complete and health risks have been eliminated. For
example, the construction of remedies to address contaminated groundwater
may be complete, but the remedies typically must operate for many years
before the site is considered cleaned up.

� EPA disagrees that the construction completion measure is not meaningful
because sites cannot be classified as construction complete until all direct
threats have been mitigated and all long-term threats brought under control.

� EPA has established the number of areas at sites where cleanup goals have
been met as an environmental indicator. Reform managers said that EPA is
considering whether it should use this indicator in addition to construction
completions.

� Views on reform's benefits/effectiveness: An industry representative
agreed that the reform has increased construction completions.

EPA reform managers said that no additional legal authority is needed to
pursue this reform. However, they predicted that program budget cuts in
fiscal year 2000 could interfere with the agency's ability to meet its
construction completion goals for fiscal year 2001 and beyond.

EPA created the National Remedy Review Board in November 1995 to review
certain proposed cleanup strategies and recommend how or whether they can be
improved. The Board reviews all proposed cleanup strategies whose estimated
costs are (1) more than $30 million or (2) more than $10 million if the
proposed strategy is 50 percent costlier than the least costly protective
alternative that complies with all appropriate environmental laws and
regulations. In fiscal year 1998, the Board also began reviewing certain
proposed removal actions estimated to cost more than $30 million. The
Board's recommendations are not binding, but regional managers must explain,
in a memorandum, how they have considered these recommendations. Table 11
shows EPA's performance goals and measures for this reform.

Table 11: Performance Goals and Measures for Establishing the National
Remedy Review Board

                                          EPA's performance measures
 Reform          Goals                    Outcomes          Outputs
                 Select more consistent
 Establish a     remedies across the
 board composed  nation                   Estimated         Number of
 of EPA managers                          savings from      cleanup
 and senior      Improve remedies'        reviewed          proposals
 technical and   cost-effectiveness       proposals: $70.7  reviewed: 43
 policy experts                           million through   through the end
 to review       Ensure that decisions    the end of the    of the first
 proposed        are in accordance with   first quarter of  quarter of
 high-cost       current laws,            fiscal year 2000  fiscal year
 cleanup actions regulations, and                           2000
                 guidance

Source: GAO's classification and presentation of information from EPA.

To date, 7 of the Board's 43 reviews have resulted in estimated savings
totaling $70.7 million, as table 12 illustrates.

Table 12: Number of Board Reviews and Estimated Savings, Fiscal Years
1996-2000

Dollars in millions

                                       1996  1997  1998  1999  2000a Total
 Number of remedies reviewed           11    9     10    11    2     43
 Number of reviews resulting in
 savings to date                       3     3     0     1     0     7
 Estimated savings to date             $34.2 34.9  0     1.6   0     $70.7

aData are for the first 3 months of fiscal year 2000.

Source: GAO's presentation of data from EPA.

The Board has reviewed cleanup strategies for sites in all 10 EPA regions
(see fig. 18).

Figure 18: Each Region's Percentage of National Remedy Review Board Reviews
Source: GAO's presentation of data from EPA.

� Sustained effects: Data from EPA suggest that this reform has resulted in
a consistent number of reviews annually since the Board was established in
fiscal year 1996. This consistency is due, at least partially, to budgetary
limitations on the number of reviews the Board can conduct each year.

� Contribution of other factors: The EPA manager for this reform said that
EPA's estimates of cost savings from the Board's reviews do not include any
savings that may have resulted from other factors.

� Regional implementation: All regions have submitted at least one proposed
cleanup strategy to the Board for review. However, because the number of
proposed remedies that qualify for the Board's review is likely to vary
among the regions, EPA's data are insufficient to determine whether the
regions are implementing the reform consistently.

� Performance measurement:

� EPA's performance measures for this reform--the number of proposals
reviewed and the dollars saved--indirectly measure progress toward two of
the reform's goals: selecting more consistent remedies and ensuring that
decisions are in accordance with current laws, regulations, and guidance.
Neither measure indicates whether the remedies have accomplished these
goals, although the Board checks for consistency and conformance during its
review process. The dollars saved measure addresses the cost component of
cost-effectiveness, but not the effectiveness component.

� Developing a mechanism to track the extent to which the Board's reviews
have led to the selection of more consistent remedies could be difficult,
particularly for sites the Board has not reviewed, as the EPA manager for
this reform acknowledges. The Inspector General's review of the Board
concluded that controls to ensure reviews of all qualifying decisions would
be difficult and costly to implement.

� Views on reform's benefits/effectiveness:

� EPA estimates that the average cost of all new Superfund cleanup actions
fell 25 percent from 1987 to 1998. EPA reform managers believe the Board's
reviews have decreased the costs of cleanups, both directly at the sites
with proposed remedies that have been reviewed and indirectly at other sites
where lessons learned from the reviews have been applied. Two industry
representatives agreed that lessons learned from the Board's reviews have
had a positive impact at other sites.

� Three industry representatives said that the potential for the Board to
review a remedy decision has led the regions to share information on
remedies and, in some instances, to push less often for what responsible
parties believe are unnecessarily expensive remedies.

� An industry representative told us that the $30 million cost threshold
excludes too many high-cost remedies from review. EPA's position is that the
threshold was chosen to result in a manageable number of reviews for the
Board, considering its available staff and financial resources. EPA believes
the Board can reasonably manage reviews of about 10 percent of Superfund
cleanup proposals. Because the actual number of reviews has fluctuated
between 9 and 13 percent, EPA has chosen not to lower the threshold.

� Industry representatives also believe the Board's 10-page limit on
responsible parties' technical submissions to the Board does not provide
adequately for addressing complex site issues. Parties would like more
direct input into the Board's reviews. In fiscal year 1997, EPA increased
the limit from 5 to 10 pages in response to concerns expressed by
responsible parties and others and believes that this increase provides for
sufficient input.

EPA reform managers said that no additional legal authority is needed to
pursue this reform.

Under the Superfund Accelerated Cleanup Model, introduced in 1992, EPA
streamlines the steps used to (1) study a site's contamination and design a
cleanup method and (2) assess the site's conditions and risks at various
stages. First, the agency conducts non-time-critical removals, or
substantial, nonemergency, shorter-term cleanup actions at portions of a
site. Second, the agency performs combined site assessments, which join the
preliminary assessment of a site with the site inspection process, and
integrated site assessments, which merge the assessments conducted before a
shorter-term removal action and the longer-term remedial actions at a site.
EPA included the model among the administrative improvements it announced in
June 1993. Table 13 shows EPA's performance goals and measures for this
reform.

Table 13: Performance Goals and Measures for Using the Superfund Accelerated
Cleanup Model

                                 EPA's performance measures
 Reform              Goals       Outcomes         Outputs
                                                  Number of
 Use the removal                                  non-time-critical,
 program to conduct  Expedite                     short-term cleanup
 nonemergency        cleanups    Time savings     actions, or removals,
 cleanup actions at              from integrating initiated: 223 through
 portions of sites   Enhance     site assessment  fiscal year 1999
 that otherwise      states'     activities
 would be addressed  roles                        Number of combined site
 by the remedial                 Cost savings     assessments performed:
 program and merge   Address     from integrating 405 through fiscal year
 assessments of      worst       site assessment  1999
 conditions and      threats     activities
 risks at selected   first                        Number of integrated site
 sitesa                                           assessments performed:
                                                  442 through fiscal year
                                                  1999

aThe model also included other components, such as efforts to encourage the
use of presumptively preferred remedies at all appropriate sites, increase
states' and communities' participation in the program, initiate enforcement
activities earlier, and address the worst threats first. These efforts,
incorporated into later reforms, are addressed in our summary of those
reforms.

Source: GAO's classification and presentation of information from EPA.

Through this reform, EPA increased the number of non-time-critical removals,
thereby expediting cleanups and saving money (see fig. 19).

Figure 19: Number of Non-Time-Critical Removals, Fiscal Years 1980-99
Source: GAO's presentation of data from EPA.

In 1995, EPA site managers estimated that non-time-critical removals, on
average, reduce the time and cost of remedial actions from 4 years and about
$4.1 million to 2 years and about $3.6 million, producing savings of 2 years
and about $0.5 million.

During fiscal years 1992-99, there was wide variation in the number of
non-time-critical removals performed across the regions, relative to each
region's share of Superfund sites (see fig. 20).

Figure 20: Each Region's Percentage of National Non-Time-Critical Removals,
Fiscal Years 1992-99, Compared With Its Percentage of National Superfund
Sites
Note: National Superfund sites include final and deleted National Priorities
List sites as of Sept. 30, 1999.

Source: GAO's presentation of data from EPA.

Assessments

From October 1, 1992 through January 10, 2000, EPA performed combined site
assessments--both the preliminary assessment of a site and the site
inspection process--at 430, or 30 percent, of the 1,421 sites that were
eligible for such assessments. From 1994 to 1999, EPA also performed 442
integrated assessments. These merge the assessments conducted before a
shorter-term removal action and the longer-term remedial actions at a site.

EPA estimates that combined assessments, on average, save about $2,500 and,
more important, shorten the cleanup process by about 11 months. EPA said
that it could not compute cost and time savings for integrated assessments
because they are used at sites that are generally larger, have higher
volumes of waste and more types of contamination, and present more complex
cleanup issues than sites that do not use such assessments. Therefore, the
two types of sites are not comparable.

The number of integrated assessments performed across the regions varied
widely relative to each region's share of Superfund sites, as figure 21
illustrates.

Figure 21: Each Region's Percentage of National Integrated Assessments,
Fiscal Years 1994-99, Compared With Its Percentage of National Superfund
Sites
Note: National Superfund sites includes final and deleted National
Priorities List sites as of Sept. 30, 1999.

Source: GAO's presentation of data from EPA.

� Sustained effects: Through this reform, EPA has increased the number of
non-time-critical removals and combined and integrated site assessments,
thereby reducing the cost and time required for site cleanup activities.
EPA's data suggest that the agency has been able to sustain the reform and
these positive effects.

� Contribution of other factors: EPA reform managers identified no other
factors as contributing to the results credited to this reform.

� Regional implementation: EPA's data and our prior reviews show that the
regions are not consistently using these cleanup tools; however, EPA has not
determined the reasons for this inconsistency, whether the reasons are
valid, or whether the agency needs to improve the reform's implementation.

� EPA's use of non-time-critical removals has been constrained, in part, by
budgetary factors. Specifically, limits on the agency's removal budget and
the higher funding priority given to emergency removals and time-critical
removals have restricted EPA's use of non-time-critical removals. However,
since 1996, the regions have been able to submit requests to a panel of
program experts from all 10 regions and EPA headquarters for funding
non-time-critical removals at Superfund sites that they could not
accommodate within their removal budgets. For the past few years, all such
removal requests have been funded.

� Statutory factors have also constrained EPA's use of non-time-critical
removals, limiting both the time and the money the agency can spend on them.
The limits are generally 12 months and $2 million per removal, although the
regions may request exemptions from these limits. In 1996, we reported that
two regions had performed very few such removals because these regions had
conservatively interpreted the statutory requirements for exemption from the
limits and, therefore, did not pursue removals or seek exemptions.8 The
regions decide whether and how many such removals to propose, and EPA
headquarters does not track consistency among the regions or determine
whether the regions are considering all possible sites for this reform.

� EPA has not tried to assess how well the regions are implementing combined
and integrated assessments or why the rate of implementation has varied so
much among the regions. The agency leaves it up to the regions to decide how
extensively to use the assessments. EPA reform managers offered several
possible explanations for the differences among regions. First, the types of
sites in the regions' workloads vary, and not all types would benefit from
combined or integrated assessments. Second, the regions take different
approaches to conducting site assessment and removal assessment activities.
Third, in some regions, the site assessment and removal programs are more
integrated, making it easier to perform combined and integrated assessments.
Finally, the percentages of sites assessed under state environmental
programs through cooperative agreements with EPA vary from region to region.

� Performance measurement:

� While EPA counts the number of non-time-critical removals initiated each
year, it does not track the dollars and time saved as measures of this
reform's progress. The EPA managers for the reform explained that the
estimates of time and cost savings are based on professional judgment rather
than data.

� EPA does track the time and cost savings from combined site assessment
activities to demonstrate progress toward this reform's goals.

� The number of removals initiated and the number of integrated and combined
site assessments performed do not directly measure EPA's progress toward or
achievement of the reform's goals--expediting cleanups, enhancing states'
roles, and addressing the worst threats first.

� Views on reform's benefits/effectiveness: Industry representatives gave
this reform considerable credit for bringing more flexibility and
cost-effectiveness into the remedy selection process. In particular, they
point to EPA's use of its removal authority for accelerating cleanups.

� Both GAO, in its prior reviews, and EPA have recognized that the current
statutory limits on the time and dollars the agency can spend on
non-time-critical removals hinder the agency's ability to use this tool more
widely for accelerating cleanups.

� In a previous review, we determined that raising the limits to at least 2
years and $4 million would allow for greater use of this tool, although
raising the limits to 3 years and $5 million would provide the maximum
flexibility.

� EPA previously asked the Congress to consider taking legislative action to
raise the statutory limits on these removal actions.

(2) Early and More Effective Community Involvement

To enhance community involvement in the cleanup process, EPA has

� since 1993, encouraged communities to form advisory groups, composed of
citizens affected by hazardous waste sites, by providing administrative
support and guidance;

� since 1988, awarded grants of up to $50,000 to eligible communities
affected by Superfund sites to enable them to acquire independent technical
assistance to help them understand and comment on site-related information
during the cleanup decision-making process; and

� since 1994, provided university educational and technical resources to
help community groups understand the technical issues at Superfund sites not
on the National Priorities List.

Table 14 shows EPA's performance goals and measures for this reform.

Table 14: Performance Goals and Measures for Promoting Community Involvement

                                        EPA's performance measures
 Reform            Goals                Outcomes          Outputs
                                                          Number of
                                                          community
 Encourage and                                            advisory groups
 assist community                                         formed: 53
 working groups at                                        through fiscal
 some sites, award                      Percentage of     year 1999
 grants to some                         surveyed
 communities near                       community members Number of
 Superfund sites                        at seven          communities
 to hire technical Enhance public       Superfund sites   receiving
 advisers, and     participation in the who believe that  technical
 provide           site decision-making EPA is            assistance
 educational and   process              effectively       grants: 219
 technical                              involving them in
 resources to some                      the Superfund     Number of
 communities near                       process: 47       communities
 Superfund sites                        percent           receiving
 not on the                                               assistance
 National                                                 through EPA's
 Priorities List.                                         Technical
                                                          Outreach Services
                                                          for Communities
                                                          program: 110

Source: GAO's classification and presentation of information from EPA.

Figures 22, 23, and 24 show EPA's efforts to enhance community involvement
through the use of three initiatives: promoting community advisory groups,
providing technical assistance grants, and conducting technical outreach
activities.

Figure 22: Community Advisory Groups Established, Through the End of Fiscal
Year 1999
Source: GAO's presentation of data from EPA.

Figure 23: Technical Assistance Grants Awarded, Fiscal Years 1988-99
Source: GAO's presentation of data from EPA.

Figure 24: Technical Outreach Projects Initiated, Fiscal Years 1994-99
Source: GAO's presentation of data from EPA.

As of November 1999, EPA had helped form 53 community advisory groups in
communities affected by Superfund sites. Forty-seven of these groups were
active at that time, and the other six, having completed their work, were no
longer active. Region V had the most sites with these groups (nine), while
Regions I and II had the fewest (two each) (see fig. 25).

Figure 25: Each Region's Percentage of National Community Involvement
Activities, Through the End of Fiscal Year 1999
Source: GAO's presentation of data from EPA.

In addition, through the end of fiscal year 1999, EPA had awarded 219
technical assistance grants totaling almost $16 million to community groups
affected by Superfund sites (only one grant is available per site). Region
IV had the most grant recipients (30), while Region VII had the fewest (4).
Finally, through the end of fiscal year 1999, EPA's outreach program had
provided technical assistance to 110 communities affected by hazardous waste
sites. Region IX provided assistance to the most communities (19), while
Region III provided assistance to the fewest (3).

� Sustained effects: The annual number of (1) new community advisory groups
increased from fiscal year 1995 through fiscal year 1997 but fell in fiscal
years 1998 and 1999, (2) technical assistance grants awarded peaked in
fiscal year 1992 at 37 but has decreased since then, and (3) new technical
outreach projects increased significantly from fiscal year 1996 through
fiscal year 1998 but fell by more than 50 percent in fiscal year 1999. These
declines may indicate that EPA is not sustaining its implementation of these
reforms.

� Contribution of other factors: According to EPA reform managers, a number
of factors affect the use of community involvement programs, including (1)
communities' level of interest in the programs; (2) the extent to which EPA
has already met communities' needs and interests; (3) the extent to which
activities that communities would become involved in, such as remedial
investigations and studies, have already been completed; and (4) the
resources available for such programs.

� Regional implementation: EPA's data show variations among the regions in
their implementation of these reforms that do not correspond with
differences in the sizes of their Superfund workloads. According to EPA
reform managers, one possible explanation for these variations is that
numerous factors can influence whether communities become involved in the
cleanup process, and these factors may vary considerably among the regions.
EPA reform managers said that although EPA can promote community
involvement, it cannot control a community's decision to participate in its
activities.

� Performance measurement: overall community involvement program:

� In fiscal year 1999, EPA completed the first phase of an ongoing effort to
measure the effects of its overall community involvement program. The review
was conducted at seven sites in four EPA regions and consisted of written
surveys, telephone surveys, and focus groups of local community members who
had shown an interest in their sites. While EPA acknowledged that the
results of the review are not statistically significant, it found that (1)
citizens perceive less risk after receiving an EPA fact sheet or attending a
public meeting about a site; (2) communities that are involved are more
likely to accept EPA's decisions and actions, ultimately making cleanups
easier, faster, and less costly; and (3) nationally, only about half of
those surveyed believe that EPA is effective in involving their communities
in the Superfund process. From this review, the agency concluded that
improvements in its community involvement efforts are needed.

� EPA is currently conducting the second phase of its performance
measurement effort, reviewing the community involvement program at 23
Superfund sites in six EPA regions. The agency expects to present the
results of these reviews in September 2000 as part of its strategic plan.

� Performance measurement: community advisory groups:

� EPA acknowledges that the growing number of community advisory groups does
not demonstrate that these groups are achieving their goals, but the agency
considers this increase an indication that communities find the advisory
groups useful.

� In fiscal year 1998, an informal review of the effectiveness of six
community advisory groups at sites in Region V found that these groups had
(1) provided an effective forum for interested parties to discuss their
views, (2) increased residents' knowledge of issues at sites, (3) often
increased the communities' voice in decisions about the sites, and (4)
improved the relationships between EPA and the communities.

� Performance measurement: technical assistance:

� EPA's Office of the Inspector General completed a review of the technical
assistance grants program in fiscal year 1996 and concluded that the 151
grants awarded through the end of fiscal year 1994 provided support for a
relatively small fraction of the 1,250 Superfund sites where community
groups were eligible to receive grants.

� The Office of the Inspector General identified possible explanations for
the limited grant activity, including the following: (1) EPA had not
assessed the number of communities that wanted grants, (2) EPA had not
effectively publicized the program, and (3) the regions had placed different
levels of emphasis on implementing and promoting the grant program.

� The number of communities that had received technical assistance grants
(219) through the end of fiscal year 1999 was small compared with the number
of Superfund sites that were on the National Priorities List (1,213).
However, EPA's technical assistance program manager told us that not all
communities need technical assistance grants, especially those whose needs
for technical information have been met by EPA's other community involvement
activities.

� Since the Office of the Inspector General completed its review, EPA has
publicized its technical assistance program on the Internet and is
developing new regional guidance intended to ensure consistent regional
implementation of the program.

� Views on reform's benefits/effectiveness:

� According to an industry representative and a state representative,
special interests often dominate community advisory groups; therefore, the
groups do not necessarily represent the needs of the community.

� According to one industry representative, communities continue to find
grant requirements overly burdensome. As a result, some communities avoid
the grant process in favor of seeking technical assistance from responsible
parties. In August 1999, EPA issued a proposed rule to further streamline
its grant application and administrative requirements.

� An industry representative also pointed out that communities often use
technical assistance grants to obtain assistance from technically
unqualified sources; therefore, EPA should enhance the qualifications for
those providing assistance. EPA's proposed rule on streamlining the grant
process includes a new requirement that technical advisers have experience
communicating problems and issues associated with hazardous or toxic waste,
redevelopment, relocation, and health to the public.

EPA reform managers said that no additional legal authority is needed to
pursue the reform.

In November 1993, EPA began providing selected states, tribes, or
municipalities with up to $200,000 each under cooperative agreements to
assess the extent and nature of any contamination at abandoned, idled, or
underutilized properties (brownfields) and plan cleanup activities. EPA
included this ongoing initiative in its administrative reforms announced in
1995. Table 15 shows EPA's performance goals and measures for this reform.

Table 15: Performance Goals and Measures for Funding Brownfield Assessment
Pilot Projects

                                       EPA's performance measures
 Reform             Goals              Outcomes          Outputs
                    Demonstrate models
                    of successful
                    brownfield         The number of
                                       properties
 Provide funds to   redevelopments thatassessed, cleaned Number of
 states, tribes,    states, tribes, andup, and           demonstration
 and municipalities localities can use redeveloped and   pilot projects
 to assess          to address         the number of     funded: 305
 contamination at   remaining          jobs and amount   through October
 brownfield         brownfields        of private        1999 (GPRA
 properties                            funding leveraged measure)
                    Through the reuse
                    of brownfields,    (data
                    encourage new jobs problematic)
                    and economic growth

Source: GAO's classification and presentation of information from EPA.

As of October 1999, EPA had funded 305 pilot projects. The recipients
voluntarily report data on the results achieved with this funding, and EPA
collects the information in its brownfield management information system.
Using this information, EPA reported, among other things, that

� 3,255 properties have been targeted for assessment, cleanup, or
redevelopment activities;

� 601 assessed properties do not require any cleanup before redevelopment;

� 120 properties have cleanup actions completed;

� 169 properties have redevelopment completed;

� about 1,617 cleanup jobs and $140 million in cleanup funds have been
leveraged at pilot properties; and

� about 4,267 redevelopment jobs and about $1.7 billion for redevelopment
activities have been leveraged.

The EPA managers for this reform also said that the pilot projects are
helping to shift incentives for development away from
greenfields--undeveloped sites in rural and suburban areas--and toward
brownfields.

� Sustained effects: EPA awarded 45 assessment pilot projects in fiscal year
1997, 105 in fiscal year 1998, and 79 in fiscal year 1999.

� Contribution of other factors:

� Contributions from other federal agencies and states have also helped to
achieve the results that EPA attributes to its administrative reform, and
EPA reform managers acknowledged this.

� In April 1999,9 we outlined funds that the Economic Development
Administration within the Department of Commerce and the Department of
Housing and Urban Development were making available to communities for
cleaning up and redeveloping brownfields.

� In April 1997, we reported that states were addressing more and more
brownfield properties through their voluntary cleanup programs.10 Under
these programs, states provide incentives, such as relief from state
liability laws, for parties so that they will voluntarily clean up and
redevelop brownfields.

� Isolating the extent to which EPA's assessment funds contributed to the
cleanup and redevelopment of the brownfield properties in the pilot projects
is not possible.

� Regional implementation: Pilot recipients are chosen on a competitive
basis by a panel consisting of EPA regional and headquarters staff and other
federal agency representatives; individual EPA regions do not decide how
many pilot projects to award. Therefore, we did not assess the level of
regional implementation of this reform.

� Performance measurement:

� EPA's Inspector General, in a March 1998 report,11 credited this
initiative with helping to remove barriers to redevelopment and leveraging
millions of dollars in private funding for redevelopment. However, the
Inspector General noted that at some of the pilot projects reviewed, EPA
funds had relatively little impact on redevelopment.

� While EPA's assessment funds played some role in achieving the results,
EPA's data are problematic. Recipients may voluntarily provide EPA with an
estimate of the number of jobs and additional funding leveraged at
properties assessed with grant funds. But the estimates are not
comprehensive because EPA cannot require the recipients to submit such data,
and EPA cannot verify the consistency and accuracy of these estimates. EPA
recently standardized the measures for which recipients report data. As a
result, the data reported by recipients may become more consistent.

� Views on reform's benefits/effectiveness: An industry representative and
cleanup officials from two states spoke positively of EPA's achievements
under this initiative.

� EPA supports legislation that would provide limited liability protection
for prospective purchasers, contiguous property owners, and innocent
landowners (those not responsible for or aware of contamination on a
property) as a means to remove liability barriers to the redevelopment of
brownfields. Cleanup officials in two states and two industry
representatives told us they also want these liability protections
established in law.

� The EPA manager for this reform also said the agency supports legislation
that would provide grants to local governments to help them fund loans for
cleaning up brownfield sites.

Characteristics of 48 Reforms That EPA Did Not Classify as Both Fundamental
and Measurable

In addition to the 14 Superfund administrative reforms discussed in appendix
I, EPA has initiated 48 other reforms. EPA reform managers classified 6 of
the 48 reforms as fundamental, meaning that they have fundamentally changed
the Superfund program. However, the agency has not been able to establish
performance measures for these reforms. The reform managers did not classify
the remaining 42 reforms as fundamental, although the agency has established
performance measures for some of these reforms.

Measurable

While reporting that they had established performance measures for 14 of the
Superfund administrative reforms that they classified as fundamental (see
app. I), EPA reform managers said they were unable to establish such
measures for 6 other reforms that they also classified as fundamental. These
included, among others, guidance to streamline the selection of cleanup
remedies and procedures for deleting cleaned portions of sites from the
National Priorities List to encourage their reuse. Table 16 lists the six
reforms and provides, for each, (1) a brief description, (2) the goals EPA
expected to achieve, (3) the types of outputs (such as specific products or
activities), (4) the effects that EPA believes cannot be measured at all or
cannot easily be measured, and (5) our observations, where possible, based
on our past work.

Table 16: Six Fundamental Reforms That Lack Performance Measures

Continued

                                                          Results and effects of
                                                        reform identified by EPA
                                          Reform's
      Reform         Description of                       Outputs    Nonmeasurable GAO's observations
                         reform                        (unverified)     effects
                                            goals
                                                                                   EPA stated that
                                                                                   "limited resources
                                                                                   prevent the agency
                                                                                   from gathering
                                                                                   property
                                                                                   transaction data
                                                                                   to quantify deals
                                                                                   facilitated by the
                                                                                   removal of sites
                                                                                   from CERCLIS."
                  Archive sites in                     Fact sheet,
                  which EPA has no                     "Archival of                GAO's report
                  further interest so  Remove the      CERCLIS                     Hazardous Waste:
                  that stakeholders,   stigma          Sites"                      Unaddressed Risks
 Refine the       such as property     associated with                             at Many Potential
 Comprehensive    owners and           inclusion in    Inventory of  Reduction in  Superfund Sites
 Environmental    purchasers, can make EPA's database  archived      perceived     (GAO/RCED-99-8,
 Response,        better decisions     of potential    sites,        potential     Nov. 30, 1998)
 Compensation, andabout properties     Superfund sites organized by  environmental recommended that
 Liability        that were once                       state and     liability at  EPA correct errors
 Information      included in CERCLIS  Restore         posted on the archived sitesin the CERCLIS
 System (CERCLIS)                      formerly        Internet                    database. The
                  Scope: Programwide   contaminated                                database included
                                       sites to                                    sites that did not
                                                       Number of
                  Status: Fully        beneficial use  archived                    meet the Superfund
                  implemented/ongoing                  sites: 31,784               program's
                                                                                   technical
                                                                                   criteria, had
                                                                                   already been
                                                                                   cleaned up, or
                                                                                   were being cleaned
                                                                                   up. In response,
                                                                                   EPA revised its
                                                                                   procedures to
                                                                                   ensure more
                                                                                   accurate CERCLIS
                                                                                   entries.
                                                                     Potential time
                                                                     and cost
                                                                     savings
                                                       Guidance
                                                       documents,    More realistic
                  Produce guidance                     including,    assumptions
                  documents on soil    Expedite        among others, about land use
                  screening, land use, cleanups        Soil
                  and presumptive                      Screening     Clearer, more
                  ("standardized")     Reduce cleanup  Guidance:     consistent    EPA is conducting
 Develop guidance remedies             costs           Users' Guide, records of    a review of the
 for remedy                                            Land Use in   decision      use of presumptive
 selection        Scope: Programwide   Promote         the CERCLA                  remedies, to be
                                       consistency in                              completed in
                                                       Remedy        Better
                  Status: Fully        remedy          Selection     understanding fiscal year 2000.
                  implemented/ongoing  selection at    Process; and  among
                                       sites
                                                       Presumptive   stakeholders
                                       nationwide      Remedies:     of EPA's
                                                       Policy and    remedy
                                                       Procedures    selection
                                                                     process and
                                                                     rationale for
                                                                     decisions
                                       Restore                       More favorable
                                       formerly        Guidance      public        EPA believes that
                                                       documents                   the benefits of
                  Delete portions of   contaminated                  perception of this reform could
                  sites from the       sites to        Number of     sites with    potentially be
                  National Priorities  beneficial use  sites with    deleted       measured by the
                  List that have been                  deleted clean portions      increase in value
 Delete clean     cleaned up and are   Allay concerns  parcels: 16                 of parcels that
 parcels from the available for        of potential                  Greater       have been deleted
 National         productive use       investors or    Number of     potential for and of surrounding
 Priorities List                       developers who  notices of    redeveloping  properties;
                  Scope: Programwide   may be          intent to     partially     however, EPA has
                                       reluctant to                  deleted sites
                                                       delete clean                not used this
                  Status: Fully        undertake       parcels from                measure because it
                  implemented/ongoing  economic        the National  Economic and  lacks data needed
                                       activity at                   other benefits
                                       these           Priorities    for the       for meaningful
                                       properties      List: 2       community     analysis.
                                                                                   According to our
                                                                                   report Superfund:
                                                                                   Progress Made by
                                                                                   EPA and Other
                                                                                   Federal Agencies
                                                                                   to Resolve Program
                                                                                   Management Issues
                  Establish the                        Panel                       (GAO/RCED-99-111,
                  National Risk-Based                  established                 Apr. 29, 1999),
                  Priority Panel,                                                  EPA uses relative
                  comprising program                   Number of                   risk to set
                  experts from all 10                  projects                    cleanup priorities
                  regions and                          reviewed in                 for sites on the
                  headquarters, to                     fiscal year                 National
                  evaluate proposed                    1997: Over 50               Priorities List.
                  cleanup actions on                                               However, EPA does
                  the basis of (1)                     Number of                   not necessarily
                  risks to humans and                  projects                    place the riskiest
                  the ecology; (2) the                 funded in                   sites on the list.
                  stability and                        fiscal year                 Many states are
                  characteristics of                   1997: 35,     Allocation of now addressing
 Promote          contaminants; and                    valued at     funding for   sites whose risks
 risk-based       (3) economic,        Better protect  $185 million  response      are severe enough
 priority-setting social, and          human health                  actions       to qualify them
 for sites on the programmatic         and the         Number of     according to  for listing. EPA
 National         considerations       environment     projects      the highest   is not including
 Priorities List                                       funded in     priorities    these sites in its
                  Fund cleanup                         fiscal year   first         priority-setting
                  projects, apart from                 1999: 25,                   because it
                  emergencies and the                  valued at                   believes that it
                  most critical                        over $100                   does not have
                  removal actions,                     million                     enough information
                  according to the                                                 on cleanup
                  priorities                           Value of                    activities at the
                  established by the                   projects                    sites. In response
                  panel                                ranked by                   to our
                                                       panel between               recommendation
                  Scope: Programwide                   August 1995                 that its regions
                                                       and March                   work with the
                  Status: Fully                        1999: Over $1               states to obtain
                  implemented/ongoing                  billion                     this information,
                                                                                   EPA has initiated
                                                                                   discussions with
                                                                                   several states on
                                                                                   sharing
                                                                                   information about
                                                                                   cleanup activities
                                                                                   at the riskiest
                                                                                   sites.
                                                       Policy
                                                       directives
                                                       and
                                                       memorandums

                                                       Number of
                                                       sites where
                                                       the reform is
                                                       being
                                                       applied: 232
                                                                     Greater
                                                       Number of     incentives for
                                                       sites where   other
                                                       EPA (1)       potentially
                                       Reduce          offered to    responsible
                  Reduce oversight of  litigation and  discuss       parties to
                  potentially          associated      oversight     cooperate and EPA believes this
 Reduce oversight responsible parties  costs           expectations  settle        reform creates
 for cooperative  that consistently                    and upcoming                opportunities for
 potentially      perform high-quality Expedite        activities    More          more efficient
 responsible      work                 cleanups        with          cooperative,  oversight, but
 parties (improve                                      potentially   less          "data collection
 oversight        Scope: Programwide   Reduce cleanup  responsible   adversarial   is time-consuming
 administration)                       costs for both  parties: 167; atmosphere    and difficult, and
                  Status: Fully        EPA and         and (2)       between EPA   the results are
                  implemented/ongoing  responsible     issued an     and           not definitive."
                                       parties         oversight     potentially
                                                       bill as       responsible
                                                       required by   parties
                                                       the           performing
                                                       settlement    cleanup work
                                                       agreement:
                                                       196 (both
                                                       actions
                                                       completed at
                                                       161 of these
                                                       sites)

                                                       Cost savings
                                                       at selected
                                                       sites
                                                       Creation of a
                  Encourage the use of Increase        Superfund Web
                  electronic tools,    com-munication  site
                  such as the          among all
                  Internet, multimedia Superfund       Development
 Improve          computers, and other stakeholders    of a home
 communication    electronic means     and improve     page for each               None
 with stakeholders                     their access to EPA region
                  Scope: Programwide   Superfund
                                       information     Number of
                  Status: Fully                        visits to
                  implemented/ongoing  Enhance public  EPA's Web
                                       participation   sites: Data
                                                       not provided

Note: Unless otherwise noted, all data presented in this table are current
as of the end of fiscal year 1999.

Source: GAO's classification and presentation of information from EPA.

EPA has not characterized the remaining 42 reforms12 as fundamental because
they (1) have not resulted in a new way of doing business for the Superfund
program; (2) are not intended for programwide implementation and have not
been integrated into the base program's operations; and/or (3) are not being
tracked as a measure for key agency or program goals:

� 14 were designed to test new concepts at selected sites, such as options
for expediting settlements with responsible parties and for encouraging
community involvement in enforcement activities;

� 9 were intended to produce guidance that would improve consistency in
remedies, risk assessments, and other aspects of the program; and

� 19 were intended to improve the overall efficiency and effectiveness of
the program's administration by, for example, establishing an ombudsman in
each region to interact with the public on Superfund issues and allowing
responsible parties to participate in risk assessments.

The 42 reforms cover a wide range of Superfund issues, including the risk
assessments that govern cleanups, the cleanup process itself, EPA's
enforcement activities, and stakeholders' involvement in cleanup decisions.
The 42 reforms also cover a diverse range of goals, such as

� increasing fairness when enforcing the program's liability laws while also
reducing litigation and its associated costs;

� expediting cleanups and reducing cleanup costs;

� better protecting human health and the environment;

� promoting consistency in the remedies selected and in the risk assessments
conducted at sites nationwide;

� encouraging stakeholders' involvement in the risk assessment process;

� enhancing communities', states', and tribes' participation in cleanup
decisions;

� preventing minority and low-income populations from bearing the burden of
pollution;

� restoring formerly contaminated sites to beneficial use; and

� improving the overall efficiency and effectiveness of the Superfund
program.

Even though EPA did not designate these reforms as fundamental, some of them
address issues that have been central to the Superfund reauthorization
debate. Our prior observations on three of these reforms, which we believe
address significant Superfund issues, follow.

As part of its administrative reforms, EPA focussed on improving
contractors' performance by implementing (1) the Superfund Long-Term
Contracting Strategy, which transferred the agency's responsibilities for
contract management from headquarters to the regions, and (2) most recently,
the Contracts 2000 Strategy, an effort to develop a set of contracts that
best meets the program's needs while using best procurement practices.
According to our report Superfund: Progress Made by EPA and Other Federal
Agencies to Resolve Program Management Issues (GAO/RCED-99-111, Apr. 29,
1999), at the time of the report, audits of Superfund contracts--EPA's
primary tool for evaluating the adequacy of contractors' policies,
procedures, controls, and performance--were being conducted more
expeditiously than in 1997. However, we found significant differences
between EPA's estimates of what cleanup work should cost and the final
contract prices for that work, indicating potential problems with the
quality of the agency's estimates. These estimates are important because
they serve as the basis for negotiating contract prices. While EPA has
established a workgroup to assess its cost-estimating procedures and
identify solutions to any problems found, the agency has undertaken similar
corrective measures in the past and has had difficulty fully implementing
and sustaining them. EPA's new Superfund contract management information
system collects cost data, but these data may not be sufficiently specific
or timely. Furthermore, we found that EPA was continuing to pay too high a
percentage of funds to contractors for program support costs, in part
because it was still retaining more contractors than it needed and paying
their overhead costs. Finally, EPA could not provide us with documentation
describing the "Contracts 2000" team's (1) overall strategy for determining
what options it would recommend that the agency adopt to address contracting
issues and (2) time frames for implementing them. We recommended that EPA
(1) develop procedures to ensure that corrective actions will improve
cost-estimates, (2) review whether the regions are consistently implementing
corrective actions, (3) identify cost-effective methods of providing
estimators with the data needed for more accurate estimates, (4) review the
number of contracts needed, and (5) ensure that the Contracts 2000
initiative results in a comprehensive strategy to improve contract
management. EPA agreed with these recommendations and plans to take a number
of actions to implement them. Among other such actions, EPA plans to (1)
develop a corrective action strategy and milestones for preparing
independent government cost estimates, (2) review the regions'
implementation of these corrective actions and the overall quality of their
cost estimates, and (3) continue to work with the regions to organize
information useful in preparing future cost estimates. In addition, EPA has
taken a variety of steps to ensure that the Superfund program has the
appropriate contracting capacity. In July 1999, the agency issued a
Contracts 2000 implementation framework documenting roles and
responsibilities for reviewing and approving the regions' implementation
plans and defining the plans' requirements.

To better recover its costs, EPA revised (1) its systems for tracking cost
recovery data and assigning priorities to cost recovery work and (2) its
methodology for calculating indirect costs--the administrative costs of
operating the program--to increase the percentage of indirect costs that are
recovered. However, according to our report Superfund: Progress Made by EPA
and Other Federal Agencies to Resolve Program Management Issues
(GAO/RCED-99-111, Apr. 29, 1999), EPA has lost the opportunity to recover
about $2 billion in indirect costs from responsible parties because the
methodology it used to calculate these costs excluded a large portion of the
total. EPA has developed a new methodology that more accurately accounts for
its administrative costs. Cost recovery program managers estimated that
using the new methodology could increase recoveries in the cases remaining
to be settled by about $629 million. As of the date of the report, the cost
recovery program had not yet implemented this new methodology because it was
awaiting approval from EPA; the Department of Justice, which litigates cost
recovery cases; and an independent accounting firm hired to review the
methodology. Subsequently, the independent accounting firm and GAO approved
the methodology. However, until EPA uses the new methodology, it will
continue to lose the opportunity to recover these funds. Furthermore, EPA
does not have a cost recovery performance measure that compares, for each
year, the amount of costs it recovers with the amount of costs it had the
potential to recover. Therefore, the agency cannot determine how well it is
performing its cost recovery activities. EPA notes that it cannot control
some factors that influence the amount of costs it can recover, such as the
percentage of cleanups with financially viable responsible parties. We
recommended that EPA apply the new indirect cost-accounting methodology as
soon as it was approved. EPA agreed to do so.

Many states have created voluntary cleanup programs, which rely on
incentives rather than enforcement orders to accomplish cleanups. EPA is
working with states and tribes to promote these programs, including
providing financial assistance to support state voluntary cleanup program
infrastructures and to promote cooperation between states, tribes, and
regions. Our report Superfund: State Voluntary Programs Provide Incentives
to Encourage Cleanups (GAO/RCED-97-66, Apr. 9, 1997) found that EPA's
authority under Superfund to ensure that cleanups are protective of human
health and the environment and the federal law's liability provisions can
deter participation in voluntary cleanup programs because potential
volunteers fear they could face expensive and indefinite cleanup liability.
Therefore, states and responsible parties would like to enter into an
agreement with EPA that would limit the federal government's future
interests at sites where voluntary cleanups have been completed. EPA has
been working with states to develop final guidance for negotiating
agreements between EPA and the states. Under these agreements, EPA would
assure volunteers that, except in limited circumstances, it generally would
not plan to take further action at sites in voluntary cleanup programs that
meet the agency's criteria for ensuring effective and protective cleanups.
In the meantime, EPA has issued an interim memorandum outlining six criteria
for voluntary programs that its regions can use to enter into agreements
with states. These criteria, while flexible, are very general and do not
clearly establish EPA's basis for determining whether a voluntary program
will qualify for an agreement in the future. We recommended that EPA work
with states to more clearly define the criteria that state voluntary cleanup
programs should meet to obtain an agreement limiting EPA's involvement at
sites, particularly in the areas of monitoring after cleanup, acceptable
oversight practices, and public participation. EPA agreed and drafted
guidance that met our recommendations. However, after many discussions with
various stakeholders, EPA concluded that there was no consensus on critical
aspects of the guidance. The agency decided, instead, that its 1996 guidance
to the regions was appropriate for future negotiations involving state
voluntary cleanup programs.

Table 17 lists the 42 reforms that EPA officials did not characterize as
having fundamentally changed the Superfund program. The table provides (1) a
brief description of each reform; (2) the goals EPA expected to achieve
through the reform; (3) the types of outputs (such as specific products or
activities) or outcomes (measurable results) that EPA uses to measure the
success of the reform; (4) where appropriate, the data measuring the effects
of the reform; and (5) any effects of the reform that EPA believes cannot be
measured or cannot easily be measured. Because of the large number of these
reforms, we did not attempt to verify their effects as identified by EPA.

Table 17: List of 42 Reforms That EPA Did Not Specifically Identify as
Having Fundamentally Changed the Superfund Program

Continued

                                                                         Results and effects of reform identified by EPA

       Reform          Description of reform        Reform's goals    Outputs (unverified)a     Outcomes      Nonmeasurable
                                                                                              (unverified)      effects
 Pilot projects (13)
                    Evaluate options for jointly
                    funding cleanup projects
                    with potentially responsible
                    parties, quantify the cost
                    implications of these                                                                   Settlement
                    options, and pilot-test                                                                 facilitated
                    demonstration projects
                    designed to identify         Increase program's                                         Number of
 Evaluate mixed     opportunities for            fairness             Number of settlements                 settlements
 funding policy     streamlining the                                  reached: 6             None identifiedpotentially
                    joint-funding                Reduce litigation                                          increased
                    decision-making process      and associated costs
                                                                                                            Equity
                    Scope: Pilots                                                                           potentially
                                                                                                            improved
                    Status: Pilots completed/
                    results incorporated into
                    program
                                                                      Number of pilot sites:
                                                                      15 in 10 regions
                    Pilot-test several
                    procedures to (1) improve                         Number of sites where
                    the quality and timeliness                        potential de minimis
                    of searches to identify                           parties (contributors
                    potentially responsible                           of small waste
                    parties, (2) make the                             volumes) were notified
                                                 Increase program's
 Pilot early        information obtained more    fairness             within 12 months of
 searches for       accessible, and (3) identify                      starting the search: 0                "Lessons
                    a larger universe of
                                                 Reduce litigation                           None identifiedlearned" applied
 potentially        potentially responsible      and associated costs Number of sites where                 programwide
 responsible partiesparties earlier in the                            other parties were
                    process                                           notified within 18
                                                 Expedite settlements
                                                                      months of starting the
                    Scope: Pilots                                     search: 5

                    Status: Pilots completed/                         Number of additional
                    results incorporated into                         parties identified
                    program                                           early in the search
                                                                      process: 150 at one
                                                                      pilot site
                                                                      Number of pilot sites:
                                                                      18 in 8 regions

                    Pilot-test revised                                Number of pilot sites
                    settlement procedures to                          where EPA achieved
                    determine the feasibility of                      early de minimis
                    (1) achieving early                               settlements: 8
                    settlements with de minimis
                    parties (contributors of                          Number of pilot sites
                    small waste volumes), (2)                         where EPA reached
                    identifying early any issues Increase program's   early ability-to-pay
                    relating to parties' ability fairness             settlements: 5
 Expedite           to pay for cleanup, and (3)                                                             "Lessons
 settlements        giving involved parties an   Reduce litigation    Number of de minimis   None identifiedlearned" applied
                    opportunity to designate     and associated costs parties with which EPA                programwide
                    others as potentially                             settled at pilot
                    responsible parties          Expedite settlements sites: 1,397 through
                                                                      the end of fiscal year
                    Scope: Pilots                                     1998

                    Status: Pilots completed/                         Amount recovered by
                    results incorporated into                         EPA in expedited de
                    program                                           minimis and
                                                                      ability-to-pay piloted
                                                                      settlements: $22.7
                                                                      million
                                                                      Number of pilot sites:
                                                                      9

                                                                      Number of pilot sites
                                                                      where EPA issued
                                                                      allocation reports: 6
                    At selected sites, test a
                    cost-allocation approach                          Number of pilot sites
                    under which potentially                           where EPA reached
                    responsible parties may                           settlements: 7
                    settle their liability on                                                Percentage of
                    the basis of their share of                       Number of pilot sites  parties at
                    the cleanup costs using (1)                       where EPA settled      pilot sites
                    a neutral party, known as an Increase program's   before allocation      indicating that
 Allocate costs     allocator, selected by the   fairness             reports were issued: 7 transaction    "Lessons
 according to       parties to allocate the                                                  costs were     learned" applied
 parties' liability costs, and (2) a nonbinding  Reduce litigation    Average time spent to  lower than     programwide
                    out-of-court allocation      and associated costs complete the           traditional
                    process                                           allocation process at  contribution
                                                                      pilot sites with       litigation
                    Scope: Pilots                                     issued reports: 20     costs: 75
                                                                      months                 percent
                    Status: Pilots ongoing/
                    results incorporated into                         Examples of legal
                    program                                           costs at pilot sites:
                                                                      $48,000 per party;
                                                                      $421,000 per site for
                                                                      EPA and Department of
                                                                      Justice staff time;
                                                                      $193,000 per site for
                                                                      the allocator
                    Test innovative approaches
                    for community involvement in                                                            Higher-quality
                    technical settlement issues,                                                            work products
                    including (1) increasing                                                                and greater
                    public involvement in                                                                   community
                    removal actions being                                                                   acceptance and
                    implemented by potentially                                                              support
                    responsible parties and (2)                       Number of pilot sites                 resulting from
                                                                      where potentially
 Encourage communityfacilitating communication                        responsible parties                   community input
 involvement in     between the potentially      Enhance public       agreed to conduct      None identified
 enforcement        responsible parties and      participation        cleanup actions or                    Some delays
                    local citizens to develop a                                                             resulting from
                    consensus on a site's future                      investigations: 13 in                 increased
                    land use                                          9 regions                             community
                                                                                                            involvement
                    Scope: Pilots
                                                                                                            "Lessons
                    Status: Pilots ongoing/                                                                 learned" applied
                    results incorporated into                                                               programwide
                    program
                    Defer placement of certain
                    sites on the National                             Guidance documents
                    Priorities List so that
                    states or tribes can oversee                      Number of states with
                                                                      which EPA has signed                  Greater
 Integrate federal, cleanup actions at these     Enhance              agreements to defer                   state/tribal
 state, and tribal  sites conducted and funded   participation of     listing of sites: 12   None identifiedparticipation
 site management    by potentially responsible   states and tribes in states in 6 regions                   and better site
                    parties                      the cleanup process
                                                                                                            management
                    Scope: Pilots                                     Number of sites
                                                                      covered by deferral
                    Status: Ongoing                                   agreements: 31
                                                                      Number of
                                                                      states/tribes with
                    Pilot test the funding of                         block grant
                    block grants to give states                       cooperative
                    and tribes greater                                agreements: 15
                    flexibility in using funds   Enhance                                                    More efficiency
                                                 participation of
 Fund state/tribal  to conduct or provide        states and tribes in Reduction in time                     and flexibility
 block grants       assistance for Superfund     the cleanup process  needed to prepare and  None identifiedin the use of
                    cleanup activities                                process paperwork: 85                 cooperative
                                                                      percent in Illinois                   agreements
                    Scope: Pilots
                                                                      Time saved by this
                    Status: Ongoing                                   regulatory change: 3
                                                                      months at one Illinois
                                                                      site

                    Increase the role of states                       Number of pilot sites:
                    and tribes in selecting the                       11 in 6 regions
                    remedies at National
                    Priorities List sites, when                       Number of sites where
                                                                      the state (or tribe)
 Involve states and possible and consistent with Enhance              has the lead role in
 tribes in the      applicable laws and          participation of     cleanup responses,
 Superfund remedy   regulations governing        states and tribes in including the remedy   None identifiedNone identified
 selection process  cleanups                     the cleanup process  selection: Evaluation
                    Scope: Pilot                                      of states' remedy
                                                                      selection data, which
                    Status: Fully                                     EPA began collecting
                    implemented/ongoing                               in fiscal year 1998,
                                                                      ongoing
                    At the regional level, test
                    the use of approaches for                                                               More favorable
                    involving community                               Number of sites where                 perception of
                                                                      EPA successfully
 Involve community  stakeholders in the                               involved the community                EPA and the
 stakeholders in theSuperfund remedy selection   Enhance public       in remedy selection: 6                Superfund
 Superfund remedy   process                      participation                               None identifiedprocess
 selection process                                                    Compendium of best
                    Scope: Pilots                                                                           Better remedy
                                                                      practices produced
                                                                                                            selection
                    Status: Fully                                     (6/99)                                process
                    implemented/completed
                    In coordination with the
                    U.S. Public Health Service,
                    establish the Medical
                    Assistance Plan to respond
                    to the health concerns of
                    underserved citizens living
                    near hazardous waste sites                        Development of
                    by providing for, among                           Superfund Medical
                    other things, (1) physicians                      Assistance Work Group
                    trained in environmental                          (SMAWG) (phased out)
 Provide training   issues and available to      Prevent
 and health         serve the affected           minority/low-income  Number of sites
 assistance to      communities, (2) medical     populations from     selected for Medical   None identifiedNone identified
 communities        testing, (3) technical       bearing the brunt of Assistance Plan
                    assistance to local agencies pollution            program testing: 1
                    and health care providers,
                    and (4) environmental health                      Amount designated for
                    education to health care                          implementing the plan
                    providers                                         at the site: $400,000

                    Scope: Pilot

                    Status: Pilot completed
                    (Reform phased out after
                    completion)
                    Fund pilot projects designed
                    to promote community
                    involvement and
                    partnerships, relying on a   Enhance public
 Promote community  coordinator in each region   participation        Number of brownfield                  Brownfield
 outreach for       to oversee brownfield pilot                       assessment pilot                      redevelopment
 brownfield         projects and initiate other  Restore formerly     projects funded: 305   None identifiedprocess affected
 redevelopment      brownfield activities        contaminated sites   through October 1999                  by citizens'
                                                 to beneficial use                                          involvement
                    Scope: Pilots

                    Status: Pilots ongoing
                    Using classroom instruction                       Funded the National
                    and hands-on work                                 Institute of
                    experience, conduct                               Environmental Health
                    interagency pilot projects                        Science's Minority
                    to train and employ                               Worker Training
                    residents of distressed                           Program                Number of
                    neighborhoods located near   Prevent                                     students
 Provide job        Superfund and brownfield     minority/low-income  Number of Super Job    trained: 75    Good will
 training and       sites and provide outreach,  populations from     Training Initiative                   fostered in
 development        curriculum, and technical    bearing the brunt of pilots: 9              Number of      communities
                    assistance to community      pollution                                   students
                    colleges located near the                         Number of grants       employed at
                    pilot sites                                       awarded for health and sites: about 25
                                                                      safety training
                    Scope: Pilots                                     programs: 20

                    Status: Ongoing
                    Underwrite the use of
                    certain promising approaches                      Number of sites that
                    for a limited number of                           participate in
                                                 Share the risks
 Use risk-sharing toprojects, agreeing to share  associated with      risk-sharing and
 encourage the      up to 50 percent of the cost implementing         eventually use the
 implementation of  of an innovative remedy if   innovative           technologies selected: None identifiedNone identified
 innovative         it fails and subsequent      technologies         Data not provided
 technology         remedial action is required
                                                                      Number of risk-sharing
                    Scope: Pilots                Expedite cleanups    agreements negotiated:
                                                                      4
                    Status: Ongoing
 Program guidance
 (8)
                    Develop remedy selection
                    rules to flag potentially
                    "controversial" cleanup                                                                 Stakeholders'
                    decisions for senior                              Guidance documents,                   confidence in
                    managers to review,                               including Rules of                    the remedy
                    providing a consolidated                          Thumb for Superfund                   selection
 Establish remedy   guide to procedures for      Promote consistency  Remedy Selection and                  process enhanced
 selection "rules ofconsultation between         in remedy selection  Consolidated Guide to  None identified
 thumb"             headquarters and the regions at sites nationwide  Consultation                          More consistency
                    on remedy selection issues                        Procedures for                        in remedy
                                                                      Superfund Response                    selection
                    Scope: Programwide                                Decisions
                                                                                                            Some costs
                    Status: Fully                                                                           reduced
                    implemented/ongoing
                    Develop documents to (1)
                    clarify the role of cost in
                    developing cleanup options                                                              Stakeholders'
                    and selecting remedies as                                                               confidence in
                    established in existing law,                                                            the remedy
                    regulation, and policy, and                                                             selection
                                                                      Fact sheet entitled                   process enhanced
 Clarify the role of(2) promote the use of       Promote consistency  The Role of Cost in
 cost               existing policies and        in remedy selection  the Superfund Remedy   None identifiedMore consistency
                    guidance to ensure           at sites nationwide
                    cost-effectiveness                                Selection Process                     in remedy
                                                                                                            selection
                    Scope: Programwide
                                                                                                            Some costs
                    Status: Fully                                                                           reduced
                    implemented/ongoing
                    Emphasize the importance of
                    maintaining appropriate
                    national consistency in the
                    Superfund remedy selection
                    process and encourage                                                                   A more
 Develop a directiveprogram managers to make     Promote consistency  Directive entitled                    predictable
 on national        full use of existing tools   in remedy selection  National Consistency                  remedy selection
 consistency in     and consultation             at sites nationwide  in Superfund Remedy    None identifiedprocess that
 remedy selection   opportunities to promote                          Selection                             more readily
                    consistency                                                                             addresses
                                                                                                            national goals
                    Scope: Programwide

                    Status: Fully
                    implemented/ongoing
                    Develop summary sheets to
                    demonstrate the context,
                    basis, and rationale for the                                                            Clearer and more
                    remedy selected at each                                                                 consistent
                    site, including the (1)                                                                 records of
                    relationship between the     Promote consistency                                        decision on
                                                                                                            selected
 Clarify informationsite's risks and response    in remedy selection  Guidance on records of                remedies
 on remedy selectionactions and (2) costs and    at sites nationwide  decision issued (8/99) None identified
                    benefits of cleanup
                    alternatives                                                                            Stakeholders'
                                                                                                            confidence in
                    Scope: Programwide                                                                      the remedy
                                                                                                            selection
                    Status: Fully                                                                           process enhanced
                    implemented/ongoing
                                                                      Guidance document
                                                                      entitled Risk
                                                                      Assessment Guidance
                                                                      for Superfund: Human
                                                                      Health Evaluation
                                                                      Manual (Part D):
                                                                      Standardized Planning,
                                                                      Reporting, and Review
                                                                      of Superfund Risk
                                                                      Assessments
                    Develop guidance that        Ensure that risk
                    establishes national         assessments are (1)  Number (and                           Better
                    criteria for the regions to  more transparent,    percentage) of new                    decision-making
                    use in planning, reporting,  clear, consistent    remedial                              for response
                                                 and reasonable; (2)
 Establish national and reviewing risk           well scoped and well investigation/                        actions at
 criteria for risk  assessments to ensure that   designed; (3) in a   feasibility study risk None identifiedSuperfund sites
 assessment         they are consistent          standard             assessments performed
                                                                      using a generic risk                  Clearer risk
                    Scope: Programwide           presentation format; assessment statement                  assessments
                                                 and (4) easier for
                                                                      of work: Data not
                    Status: Fully                decision-makers at   provided                              Resources saved
                    implemented/ongoing          Superfund sites to
                                                 review
                                                                      Number (and
                                                                      percentage) of new
                                                                      remedial
                                                                      investigation/
                                                                      feasibility study risk
                                                                      assessments performed
                                                                      using standard tables
                                                                      for reporting risk
                                                                      data: Data not
                                                                      provided
                    Update and standardize risk
                    assessment guidance,
                    especially in areas where    Ensure the quality,
                    science and policy have      consistency, and
                    advanced over the past       reliability of risk  Guidance documents
                                                                      under development
 Develop guidance toseveral years, such as       assessments
 standardize risk   exposure assessment, human                        Meetings with          None identifiedNone identified
 assessments        health toxicity assessment,  Promote greater      stakeholders to seek
                    and risk communication       community
                                                 involvement in       input on priorities
                    Scope: Programwide           designing risk       for improvement
                                                 assessments
                    Status: Not yet fully
                    implemented/ongoing
                    Issue guidance on
                    establishing appropriate
                    cleanup levels and levels of
                    concern (soil screening
                    levels) for common chemical  Reduce time and
                    contaminants in soil and     costs of cleanups                                          Development of a
 Develop soil       complete a pilot study of                         Soil screening                        useful tool for
 screening levels   soil screening levels at 10  Promote more         guidance               None identifiedinitially
                    sites                        effective and                                              assessing a
                                                 consistent cleanups                                        site's risks
                    Scope: Programwide           nationwide

                    Status: Fully
                    implemented/ongoing
                    Use public-private
                    partnerships to demonstrate
                    and evaluate innovative
                    hazardous waste treatment                         Number of
                    technologies and increase                         partnerships: 5
                    their acceptance by (1)                           active/2 inactive or
                    targeting contamination                           completed
                    problems affecting both
                    public and private sites,    Better protect human Number of technology                  Better
                                                 health and the                                             information for
 Promote the use of (2) evaluating technology    environment          cost and performance                  decisionmakers
 innovative         databases, and (3) trying to                      case studies: 82 by    None identifiedabout
 technology         improve the dissemination of Expedite cleanups    EPA; 140 by the                       appropriate
                    information on treatment                          Federal Remediation
                    technologies to common data                       Technologies                          remedies for
                    repositories                 Reduce cleanup costs Roundtable                            sites

                    Scope: Programwide                                Remediation technology
                                                                      databases established:
                    Status: Ongoing--effort                           4
                    adopted as an
                    "administrative improvement"
                    in 1993
 Replaced/superseded
 reforms (2)
                    Issue guidance (1)
                    authorizing the regions to
                    identify uncontaminated land                      Completion of the                     More favorable
                    parcels on or adjacent to                         Federal Register                      public
                    National Priorities List                          notice and                            perception of
                    sites to facilitate the                           implementing guidance                 sites with
                    transfer, development, or                                                               deleted parcels
                    redevelopment of these                            Number of sites from
 Clarify National   parcels; (2) setting forth   Restore formerly     which clean parcels                   Greater
 Priorities List    the factual basis for        contaminated sites   had been deleted as of None identifiedpotential for
 sites              assurances that parcels are  to beneficial use    March 1999: 16                        redevelopment of
                    not contaminated; and (3)                                                               partially
                    specifying the consultation                       Number of sites for                   deleted sites
                    and coordination required                         which notices of
                                                                      intent to delete clean                Positive
                    Scope: Programwide                                parcels had been                      economic and
                                                                      issued as of March                    other effects on
                    Status: Fully                                     1999: 2                               the communities
                    implemented/ongoing
                    Give private parties an
                    opportunity to seek EPA's
                    approval of an allocation
                    that covers 100 percent of                                                              Transaction
                    the costs at a site, with    Facilitate                                                 costs for all
                    the understanding that an    settlements          Number of sites where                 parties reduced
                    approved allocation can                           private-party
 Adopt private-partyserve as the basis for a     Reduce litigation    allocations have been                 Settlements
 allocations        settlement                   and transaction      adopted: 3 settlements None identifiedfacilitated
                                                 costs                before this reform was
                    Scope: Programwide                                merged with the orphan                Parties'
                                                 Increase program's   share reform                          concerns about
                    Status: Fully                fairness                                                   fairness
                    implemented/completed                                                                   addressed
                    (incorporated into orphan
                    share compensation reform)
 Other reforms (19)
                    Issue guidance on
                    presumptive remedies
                    (standardized remedies for
                    certain types of sites),                          Guidance documents,                   More effective
                    including those for          Expedite cleanups    including Presumptive                 and consistent
 Streamline/expeditemunicipal landfills and      Promote more         Remedy for CERCLA                     remedy selection
 the cleanup processvolatile organic compounds   effective and        Municipal Landfill     None identified
                    in soil                                           Sites and Users' Guide                Time and/or
                                                 consistent cleanups
                                                                      for the VOCs in Soil                  costs of
                    Scope: Programwide           nationwide           Presumptive Remedy                    cleanups reduced

                    Status: Fully
                    implemented/ongoing
                    Issue guidance and make
                    information available to
                    address property owners'                          Guidance and other
                    concerns about, among other                       documents, including
                    issues, federal liens on                          Guidance on Agreements
                    contaminated property and                         with Perspective
                    the potential liability of                        Purchasers of
                                                                      Contaminated Property
 Promote greater    prospective purchasers of    Increase program's   and a model agreement
 fairness for       Superfund sites              fairness
 Superfund site                                                       Increase through       None identifiedNone identified
 owners             Scope: Programwide           Reduce litigation    fiscal year 1998 in
                                                 and associated costs
                    Status: Fully                                     the number of
                    implemented/ongoing                               prospective purchaser
                    (Reform partially replaced                        agreements following
                    by a later reform intended                        the issuance of the
                    to remove liability barriers                      guidance and model
                    through agreements with                           agreement: 100
                    prospective purchasers)
                    Direct staff to evaluate all
                    decision documents for the
                    possibility of
                    disproportionate adverse
                    effects on minority and
                    low-income communities in an Prevent              Policy directive
                    effort to ensure that all    minority/low-income
 Implement an       waste programs treat         populations from     Consultation with the
 environmental      environmental justice as an  bearing the brunt of National Environmental None identifiedNone identified
 justice strategy   integral part of EPA's       pollution            Justice Advisory
                    policies, guidance, and                           Council during policy
                    regulations                  Expand meaningful    development
                                                 public participation
                    Scope: Programwide

                    Status: Fully implemented/
                    ongoing
                    Encourage states,
                    territories, and tribes to
                    clean up contaminated sites
                    under their own laws by, for
                    example, working with state
                    associations to develop
                    criteria for deferring sites
                    to states, initiating pilot
 Defer certain site deferral projects in         Enhance
 categories to      qualified states, and        participation of     Guidance document      None identifiedNone identified
 states             establishing a workgroup to  states and tribes in
                    address deferral questions   the cleanup process
                    and assess early state-led
                    cleanups

                    Scope: Programwide

                    Status: Fully implemented/
                    ongoing
                    Improve contractors'
                    performance by implementing
                    (1) the Superfund Long-Term
                    Contracting Strategy, which
                    shifts responsibility for
                    contracts and contract                            Reduction in program
                    management from headquarters                      support costs,
                    to the regions, and, (2)                          measured as a
                    most recently, the Contracts Improve program's    percentage of total
 Improve contract   2000 Strategy to develop a   efficiency and       invoice costs, over
 management         set of contracts that best   effectiveness        time: Data not         None identifiedNone identified
                    meet the program's needs                          provided
                    while using best procurement Reduce cleanup costs
                    practices                                         Long-Term Contracting
                                                                      Strategy; Contracts
                    Scope: Programwide                                2000 Strategy (2/98)

                    Status: Fully implemented/
                    ongoing--effort adopted as
                    an "administrative
                    improvement" in 1993
                    Work with the Department of
                    Defense to accelerate
                    cleanup work at military                                                 Project work
                    bases designated for closure                                             (time)         Better community
                    or realignment and to                                                    reductions:    and interagency
                    address property transfer,                                               over 250 work  working
                    redevelopment, and community Expedite cleanups                           years through  relationships
 Accelerate cleanup involvement issues                                Input to Department of fiscal year
 at bases designated                             Restore formerly     Defense guidance and   1998           Better
 for closures       Scope: Programwide (108      contaminated sites   joint policy documents                relationships
                    installations)               to beneficial use                           Project costs  between the
                                                                                             avoided: $250  Department of
                    Status: Fully                                                            million throughDefense and
                    implemented/ongoing--effort                                              fiscal year    states
                    adopted as an                                                            1998
                    "administrative improvement"
                    in 1993
                    Issue regional
                    compliance-monitoring
                    guidance and implement
                    regional compliance-tracking
                    systems to strengthen
                    enforcement through          Improve program's    Regional
                    oversight of potentially     efficiency and       compliance-monitoring
 Improve compliance responsible parties'         effectiveness        guidance
 monitoring         compliance                                                               None identifiedNone identified
                                                 Better protect human Regional
                    Scope: Programwide           health and the       compliance-tracking
                                                 environment          systems
                    Status: Fully implemented/
                    ongoing--effort adopted as
                    an "administrative
                    improvement" in 1993

                    Improve systems for tracking                                             Percentage of
                    cost recovery data and for                                               targeted sites
                    assigning priorities to cost                                             valued at more
                    recovery work and revise                                                 than $200,000
                    indirect cost accounting                                                 that were
                    methodologies to increase                                                addressed in
 Improve the        the percentages of indirect                       Results of             fiscal year
 effectiveness of   costs that are recovered     Reduce cleanup costs cost-recovery planning 1998: 100      None identified
 cost recovery                                                        and targeting efforts
                    Scope: Programwide                                                       Costs recovered
                                                                                             through
                    Status: Fully implemented/                                               settlements
                    ongoing--effort adopted as                                               through fiscal
                    an "administrative                                                       year 1998:
                    improvement" in 1993                                                     About $2.4
                                                                                             billion
                    Encourage the regions to
                    consider means of
                    streamlining the grant
                    process, such as providing
                    advance funding for
                    technical assistance grants
                    and authorizing training for
                    grant recipients, to make                         Number of technical
                    resources available for                           assistance grants
                    communities to acquire                            awarded since 1988:
 Fund technical     independent technical        Enhance public       Over 202
 assistance grants  assistance that could help   participation                               None identifiedNone identified
                    them understand and comment                       Publication in the
                    on information related to                         Federal Register of a
                    Superfund sites (e.g.,                            rule to streamline the
                    records of decision on                            grant process, 8/24/99
                    cleanup remedies)

                    Scope: Programwide

                    Status: Not yet fully
                    implemented
                    Reduce the risks associated
                    with using innovative
                    technologies by providing
                    indemnification coverage for
                    the prime contractor as well
                    as the innovative technology                      Policy directive
 Use risk-sharing tocontractor, thereby
 encourage the use  protecting both from                              Number of requests for
 of innovative      third-party liability claims Expedite cleanups    expanded               None identifiedNone identified
 technologies       if the technology does not                        indemnification
                    perform as expected                               coverage received: 0

                    Scope: Programwide

                    Status: Fully
                    implemented/ongoing
                    Work with states and tribes
                    to promote programs that                          Number of states that
                    encourage private parties to                      have implemented
                    voluntarily clean up                              voluntary cleanup
                    contaminated sites by, for                        programs: 44
                    example, (1) providing
                    financial assistance to                           Number of states that
                    support an infrastructure    Enhance              have signed agreements                Ability of
                    for state voluntary cleanup  participation of     with EPA regions on                   state/tribal
                    programs and to promote      states and tribes in supporting voluntary                  programs to
 Support voluntary  cooperation among states,    the cleanup process  cleanups and                          assess, clean
 cleanup programs   tribes, and regions, and (2)                      redeveloping           None identifiedup, and recover
                    issuing guidance on drafting Restore formerly     brownfields: 14 as of                 costs at
                    memorandums of agreement     contaminated sites   December 1999                         brownfield sites
                    between EPA regions and      to beneficial use                                          enhanced
                    states in support of                              Amount distributed
                    voluntary cleanup programs                        since 1997 to support
                                                                      an infrastructure for
                    Scope: Programwide                                state/ tribal
                                                                      voluntary cleanup
                    Status: Fully                                     programs: Over $29
                    implemented/ongoing                               million

                                                                      Reference documents,   Number (and
                                                                      including Risk         percentage) of
                                                                      Assessment Guidance    new remedial
                    Develop a reference document                      for Superfund, Volume  investigation/
                    to support and promote                            1, Human Health        feasibility
                    public participation in the  Encourage            Evaluation Manual      study starts inRemedies
 Encourage communityrisk assessment process      stakeholders'        (Supplement to Part    which the      improved
 participation in                                involvement in the   A): Community          community has
 risk assessments   Scope: Programwide           risk assessment      Involvement in         been           Community
                                                 process              Superfund Risk         substantively  relations
                    Status: Fully                                     Assessments            involved in    improved
                    implemented/ongoing                                                      designing the
                                                                      Video and brochure on  risk
                                                                      citizens' involvement  assessment:
                                                                      in risk assessments    Data not
                                                                                             provided
                    Issue a directive (1)
                    confirming that responsible                                              Number (and
                    and qualified parties can                                                percentage) of
                    perform risk assessments at                       Policy document:       new remedial
                    most sites and (2) removing  Encourage            Revised Policy on      investigation/
 Allow potentially  the requirement that the     stakeholders'        Performance of Risk    feasibility
 responsible partiesregions consult with         involvement in the   Assessments During     study starts
 to perform risk    headquarters before allowing risk assessment      Remedial               where the risk None identified
 assessments        these parties to perform     process              Investigations/        assessment was
                    risk assessments                                  Feasibility Study      performed by
                                                 Expedite cleanups    (RI/FS) by Potentially the potentially
                    Scope: Programwide                                Responsible Parties    responsible
                                                                                             parties: Data
                    Status: Fully                                                            not provided
                    implemented/ongoing
                                                                      Technical Review
                    Establish an expert                               Workgroup and Lead                    More
                    workgroup to standardize                          Sites Workgroup                       opportunities
                    risk assessment approaches                                                              for the public
                    for lead-contaminated        Better protect human Guidance document:                    to interact with
                    Superfund sites and provide  health and the       Revised Interim Soil                  EPA
 Establish an expertadvice to regional risk      environment          Lead (PB) Guidance for
 workgroup on lead  assessors and site managers                       CERCLA Sites and RCRA  None identifiedFewer questions
                                                 Make risk            Corrective Action
                                                                                                            raised on
                    Scope: Programwide           assessments more     Facilities                            consistency
                                                 consistent
                    Status: Fully                                     Other guidance                        Use of better
                    implemented/ongoing                               documents and                         science advanced
                                                                      site-specific
                                                                      consultations
                    Develop guidance promoting
                    the concept of a single
                    regulator for federal sites,
                    specifying roles and
                    outlining the general
                    principles and guidelines
                    that federal and state
                                                                                                            Less duplication
 Establish a lead   partners should follow in    Improve program's                                          of effort and
 regulator for      overseeing cleanup           efficiency and       Guidance documents     None identifiedinefficiency in
 federal sites      activities, thereby          effectiveness                                              the use of
                    simplifying the cleanup
                    process and allowing for                                                                resources
                    more efficient staffing

                    Scope: Programwide

                    Status: Fully
                    implemented/ongoing
                    Revise guidance to (1)                            Amended policy
                    ensure that response actions                      document: Revised
                    taken at sites up to the                          Hazard Ranking System:
                    time of their placement on                        Evaluating Sites After
                    the National Priorities List                      Waste Removals
                    are considered in
                    determining whether the                           Number of sites                       Incentives
 Consider the       sites qualify for the list                        considered as low                     provided for
 response actions   and (2) incorporate greater                       priorities for                        parties to
 taken at sites     flexibility in evaluating    Reduce litigation    placement on the                      conduct early
 before placing themwhether sites should be      and associated costs National Priorities    None identifiedresponse actions
 on the National    removed from the list to                          List or classified as                 before sites are
 Priorities List    encourage early cleanup                           "no further remedial                  placed on the
                    actions, especially by                            action planned"                       National
                    private parties                                   because prior cleanup                 Priorities List
                                                                      activities, such as
                    Scope: Programwide                                waste removal, were
                                                                      considered in setting
                    Status: Fully                                     priorities: Data not
                    implemented/ongoing                               provided
                    Develop guidance for the
                    regions that will address
                    (1) the role of risk and
                    other factors (such as cost,
                    community concerns,                               Interim final guidance
                    environmental justice, and
                    cultural considerations) in                       Number of federal
                    setting priorities at                             facility agreements                   Three regions
                                                                                                            (III, IX, and X)
 Promote risk-based federal facilities, (2) the                       revised to reflect                    assisted in
 priority-setting   Department of Defense's and  Better protect human changes in priorities                 setting
 for federal        the Department of Energy's   health and the       within Department of   None identifiedrisk-based
 facilities         approaches to evaluating     environment          Defense and Department                priorities at
                    risks at sites, and (3) the                       of Energy facilities
                    appropriate role of                               (number of agreements                 Navy Superfund
                    stakeholders in setting                           and number of                         sites
                    priorities                                        milestones revised):
                                                                      Data not provided
                    Scope: Programwide

                    Status: Not yet fully
                    implemented/ongoing
                                                                      For each unilateral
                                                                      administrative order
                                                                      issued, the number of
                                                                      parties identified at
                                                                      the site, the number
                                                                      of parties excluded,
                                                                      and the documentation
                                                                      of reasons for
                                                                      exclusion: Data not
                    Issue unilateral                                  provided (enforcement
                    administrative orders for                         data confidential)
                    site cleanups to the largest                                                            Regions'
                    manageable number of                              Number of orders                      willingness to
                    parties, after considering                        independently reviewed                issue unilateral
                    evidence of the parties'                          by EPA headquarters to                administrative
                    liability, financial                              ensure that they had                  orders to a
                                                                                                            larger number of
 Ensure more        viability, and contribution  Increase program's   been issued to all                    potentially
 equitable issuance to a site's waste, and       fairness             appropriate parties                   responsible
 of unilateral      establish procedures                              (including             None identifiedparties
 administrative     requiring regional staff to  Reduce litigation    governmental                          increased
 orders             document their reason(s) for and associated costs entities): 180
                    proposing that certain
                    parties be excluded from                          Number (and                           Parties'
                    administrative orders                             percentage) of orders                 perception of
                                                                      that have required                    EPA regions'
                    Scope: Programwide                                documentation: Data                   fairness in
                                                                      not provided                          issuing orders
                    Status: Fully                                     (enforcement data                     improved
                    implemented/ongoing                               confidential)

                                                                      Number (and
                                                                      percentage) of cases
                                                                      where reasons cited
                                                                      for excluding parties
                                                                      were consistent with
                                                                      EPA's policy: Data not
                                                                      provided (enforcement
                                                                      data confidential)
                    Establish an ombudsman in
                    each region to facilitate
                    the resolution of regional
                    issues or problems by                                                    Number of cases
                    serving as a point of                                                    for which EPA
                    contact for the public and                                               conducted      Public's
 Establish ombudsmenhelping to resolve           Enhance public       Appointed an ombudsman investigations perceptions of
                    stakeholders' concerns       participation        in each region         and mediations:EPA's decisions
                                                                                             Data not       improved
                    Scope: Programwide                                                       provided

                    Status: Fully
                    implemented/ongoing

a Unless otherwise noted, all data are current as of the end of fiscal year
1999.

Source: GAO's classification and presentation of information from EPA.

Objectives, Scope, and Methodology

Our overall objective was to determine the effectiveness of the 62
administrative reforms EPA has introduced to improve the Superfund program.
Specifically, we sought to (1) determine the demonstrated results of these
reforms and evaluate the quality of the measures the agency uses to gauge
the results and (2) identify legislative changes to the program that either
the agency or key stakeholders--including, among others, representatives of
parties responsible for cleanups, environmental groups, and states--believe
are still necessary. In determining the scope of our review, we asked the
agency to identify those reforms that it considers to be key to the
Superfund program. The agency identified 14 of the 62 reforms that it
considers to have fundamentally and measurably changed the program, and we
focused our detailed audit work on them. This report primarily summarizes
the results of our review of the 14 reforms as a group. However, we also
provide information on the remaining 48 reforms (see app. II).

To obtain information on the 62 reforms, we developed a set of questions on
their nature, characteristics, scope of implementation, performance
measures, and results. We submitted these questions to the Senior Process
Manager for Reforms in the Office of Emergency and Remedial Response within
EPA's Office of Solid Waste and Emergency Response and to the principal
reform manager in the Office of Site Remediation Enforcement in EPA's Office
of Enforcement and Compliance Assurance at the agency's headquarters in
Washington, D.C. These reform managers coordinated responses throughout the
agency and provided us with detailed information on each of the reforms. We
also reviewed EPA's annual reports on the reforms and various information
and documents relating to the reforms available on EPA's Superfund Web site.

To determine the overall effects of key reforms on the Superfund program, we
first asked these reform managers to identify the reforms that they regarded
as "fundamental" or otherwise significant in some way. We also asked them to
explain why they classified these reforms as fundamental or significant. In
their responses, the reform managers identified 20 reforms that they
considered to have fundamentally changed the Superfund program. According to
these reform managers, EPA regards a reform as having produced a
"fundamental change" if it (1) results in a new way of doing business for
the Superfund program; (2) is intended for programwide implementation and
has been integrated into the base program's operations; and (3) where
appropriate, is being tracked as a measure for key agency or program goals.
We also asked the reform managers to identify the fundamental reforms that
have produced measurable (quantifiable) outcomes. Of the 20 reforms, these
reform managers characterized 14 as having produced measurable outcomes, and
we focused our review of the reforms' effects on the program on these 14.

To obtain the information we needed on these 14 reforms, we submitted
additional detailed questions on each reform to the agency's reform
managers. In these questions, we asked for further information on (1) each
reform's goals, (2) the extent to which EPA regions are implementing the
reform, (3) the performance measures EPA uses to track progress toward
achieving the reform's goals, and (4) the results of each reform. We also
asked EPA to provide supporting data and documentation to verify this
information. To determine the extent of the reforms' implementation and
their effects, we reviewed and analyzed the information provided by EPA. In
addition, we reviewed and synthesized information from key published and
internal EPA documents, as well as reports, studies, and analyses by other
organizations that have examined the effects of the reforms. We did not
attempt to verify the data and analyses provided. Furthermore, we
interviewed representatives of various industry, environmental, and
government groups, including, for industry groups, the Chemical
Manufacturer's Association, the Superfund Settlements Project, the National
Association of Realtors, and the National Federation of Independent
Business; for environmental groups, Environmental Defense, the U.S. Public
Interest Research Group, and Resources for the Future; and for state and
local government groups, the Association of State and Territorial Solid
Waste Management Officials, the Committee for the National Institute for the
Environment, the National Association of Counties, the National Association
of Local Government Environmental Professionals, the National Governors
Association, the U.S. Conference of Mayors, and the congressionally
chartered National Academy of Public Administration. We also talked to staff
in EPA's Office of the Inspector General and reviewed a number of the
Office's reports on Superfund. In addition, in commenting on some of the
reforms, we relied on our past work on a variety of Superfund issues.

To identify changes to the Superfund program that EPA and other stakeholders
believe are still necessary, we reviewed and analyzed EPA's written
responses to the questions we submitted on each reform. We also talked to
representatives from the industry, environmental, and state and local
government groups identified above to discuss their views on changes needed
in the Superfund program.

We conducted our work for this review between July 1999 and May 2000 in
accordance with generally accepted government auditing standards.

Comments From the Environmental Protection Agency

GAO Contacts and Staff Acknowledgments

David G. Wood (202) 512-6111

Eileen Regan Larence (202) 512-6510

In addition to those named above, Vincent P. Price, David Goldstein, and
Michael Ibay made key contributions to this report.

(160500)

Table 1: Fourteen Fundamental/Measurable Reforms and Their Output and
Outcome Measures 10

Table 2: Performance Goals and Measures for Encouraging the Use of
Alternative Dispute Resolution 24

Table 3: Performance Goals and Measures for Promoting "Enforcement First" 28

Table 4: Performance Goals and Measures for Promoting Settlements With Small
Waste Contributors 31

Table 5: Performance Goals and Measures for Reaching Agreements With
Prospective Purchasers 35

Table 6: Performance Goals and Measures for Providing Compensation for
Orphan Shares 39

Table 7: Performance Goals and Measures for Encouraging the Use of
Site-Specific Accounts 43

Table 8: Performance Goals and Measures for Revised Guidance on Settlements
With Miniscule Waste Contributors 47

Table 9: Performance Goals and Measures for Updating Remedy Decisions 49

Table 10: Performance Goals and Measures for Increasing Construction
Completions 53

Table 11: Performance Goals and Measures for Establishing the National
Remedy Review Board 57

Table 12: Number of Board Reviews and Estimated Savings, Fiscal Years
1996-2000 58

Table 13: Performance Goals and Measures for Using the Superfund Accelerated
Cleanup Model 61

Table 14: Performance Goals and Measures for Promoting Community Involvement
67

Table 15: Performance Goals and Measures for Funding Brownfield Assessment
Pilot Projects 74

Table 16: Six Fundamental Reforms That Lack Performance Measures 79

Table 17: List of 42 Reforms That EPA Did Not Specifically Identify as
Having Fundamentally Changed the Superfund Program 88

Figure 1: Comparison of Four Regions' Superfund Workloads and Rates of
Implementation for Three Reforms 15

Figure 2: Superfund Liability Cases Initiated Using Alternative Dispute
Resolution, Fiscal Years 1993-98 25

Figure 3: Each Region's Percentage of National Superfund Cases in Which
Alternative Dispute Resolution Was Used, Fiscal Years 1993-98 26

Figure 4: Percentage of New Cleanup Actions Funded by Responsible Parties,
Fiscal Years 1987-99 29

Figure 5: Average Percentage of New Cleanups Funded by Responsible Parties,
by Region, Compared With the National Average, Fiscal Years 1995-99 30

Figure 6: Number of De Minimis Settlements, Fiscal Years 1987-99 32

Figure 7: Each Region's Percentage of National De Minimis Settlements,
Fiscal Years 1987-99 33

Figure 8: Number of Prospective Purchaser Agreements Completed, Fiscal Years
1989-99 36

Figure 9: Each Region's Percentage of National Prospective Purchaser
Agreements, Fiscal Years 1989-99 37

Figure 10: Number of Orphan Share Compensation Offers, Fiscal Years 1996-99
40

Figure 11: Each Region's Percentage of National Orphan Share Compensation
Offers, Fiscal Years 1996-99 41

Figure 12: Number of Special Accounts Established, Fiscal Years 1990-99 44

Figure 13: Each Region's Percentage of National Special Accounts
Established, Fiscal Years 1990-99, Compared With Its Percentage of National
Superfund Sites 45

Figure 14: Number of Cleanup Remedy Updates per Year, Fiscal Years 1982-99
50

Figure 15: Each Region's Percentage of National Cleanup Remedy Updates,
Fiscal Years 1996-99, Compared With Its Percentage of National Superfund
Sites 51

Figure 16: Number of Sites Designated as Construction Complete, Fiscal Years
1981-99 54

Figure 17: Each Region's Percentage of National Construction Complete Sites,
Fiscal Years 1981-99, Compared With Its Percentage of National Superfund
Sites 55

Figure 18: Each Region's Percentage of National Remedy Review Board Reviews
58

Figure 19: Number of Non-Time-Critical Removals, Fiscal Years 1980-99 62

Figure 20: Each Region's Percentage of National Non-Time-Critical Removals,
Fiscal Years 1992-99, Compared With Its Percentage of National Superfund
Sites 63

Figure 21: Each Region's Percentage of National Integrated Assessments,
Fiscal Years 1994-99, Compared With Its Percentage of National Superfund
Sites 64

Figure 22: Community Advisory Groups Established, Through the End of Fiscal
Year 1999 68

Figure 23: Technical Assistance Grants Awarded, Fiscal Years 1988-99 69

Figure 24: Technical Outreach Projects Initiated, Fiscal Years 1994-99 70

Figure 25: Each Region's Percentage of National Community Involvement
Activities, Through the End of Fiscal Year 1999 71
  

1. Responsible parties include present (and some former) site owners,
operators, transporters, and persons who arrange for the treatment or
disposal of hazardous substances.

2. EPA defines brownfields as abandoned or underused facilities, usually in
industrial or commercial areas, where redevelopment is hampered by real or
perceived environmental contamination.

3. Superfund: Information on EPA's Administrative Reforms (GAO/RCED-97-174R,
May 30, 1997).

4. GAO and EPA have, in the past, disagreed on whether the methodology that
EPA uses as a basis for saying that the program is cleaning up sites faster
is appropriate (see Superfund: Times to Assess and Clean Up Hazardous Waste
Sites Exceed Program Goals (GAO/T-RCED-97-69, Feb. 13, 1997), Superfund:
Times to Complete the Assessment and Cleanup of Hazardous Waste Sites
(GAO/RCED-97-20, Mar. 31, 1997), and Superfund: Duration of the Cleanup
Process at Hazardous Waste Sites on the National Priorities List
(GAO/RCED-97-238R, Sept. 24, 1997)). We have not assessed the agency's
estimate of cost savings.

5. Superfund: Progress Made by EPA and Other Federal Agencies to Resolve
Program Management Issues (GAO/RCED-99-111, Apr. 29, 1999), and Superfund:
Half the Sites Have All Cleanup Remedies in Place or Completed
(GAO/RCED-99-245, July 30, 1999).

6. State voluntary cleanup programs offer parties incentives, such as state
liability protection, to voluntarily address waste sites.

7. Including, among others, the Alternative Dispute Resolution Act of 1998
(P.L. 105-315) and the Civil Justice Reform Act (P.L. 101-650).

8. Superfund: Non-Time-Critical Removals as a Tool for Faster and Less
Costly Cleanups (GAO/T-RCED-96-137, Apr. 17, 1996).

9. Environmental Protection: Agencies Have Made Progress in Implementing the
Federal Brownfield Partnership Initiative (GAO/RCED-99-86, Apr. 9, 1999).

10. Superfund: State Voluntary Programs Provide Incentives to Encourage
Cleanups (GAO/RCED-97-66, Apr. 9, 1997).

11. Brownfields: Potential for Urban Revitalization, EPA, Office of the
Inspector General (E1SHF8-11-0005-8100091, Mar. 27, 1998).

12. One of the reforms in the group of 14--addressing options for private
party allocations of cleanup liability--and 1 of the reforms in the group of
9--developing guidance on how to address uncontaminated parcels on or
adjacent to Superfund sites--were included in the first round of reforms but
were subsequently replaced or incorporated into the later two rounds.
*** End of document. ***