SIJPERFIJND RD/RA Negotiation Time Frames E1SFG7-11-0024-8400015 March 27, 1998 ------- Inspector General Division Headquarters Audit Division Conducting the Audit Washington, DC Regions Involved Regions 3 and 5 Program Office Involved Office of Enforcement and Compliance Assurance, Office of Site Remediation Enforcement ------- ^tDSX $ Si % uffiy UNITED STATES ENVIRONMENTAL PROTECTION AGENCY WASHINGTON, D.C. 20460 MAR 2 7 1998 OFFICE OF THE INSPECTOR GENERAL SUBJECT: RD/RA Negotiation Time Frames Report No. E1SFG7-11-0024-8400015 FROM: Norman I Divisional Inspector General for Audit Headquarters Audit Division TO: Barry Breen, Director Office of Site Remediation Enforcement Office of Enforcement and Compliance Assurance Linda Boornazian, Director Policy and Program Evaluation Division Office of Site Remediation Enforcement Office of Enforcement and Compliance Assurance Attached are the results of our review of time frames for negotiating remedial design and remedial action (RD/RA). This report is designed to provide you information as part of the OIG's Advisory and Assistance Services. We, therefore, have not included formal recommendations. We have included several suggestions for your consideration. This concludes our work on this assignment and no written response is necessary We analyzed overall trends in the length of RD/RA negotiations for fiscal years 1990-1996 and whether Superfund enforcement reforms affected the length of negotiations. In general, we found the average time frames for RD/RA negotiations were longer in later years (1995 and 1996) than in earlier years. We also noted that changes in the time frames coincided with two major policy changes, "enforcement first" and the Superfund reforms. In addition, different regional management approaches appear to have influenced RD/RA negotiation time frames. Case specific factors, which we did CJi.Recycled/Recyclable Printed with Soy/Canola Ink on paper that aC-7 contain# at least 50% recycled fiber ------- not include in our analysis, and many of which are not within EPA's control, can also affect RD/RA negotiation time frames. You may want to consider the following suggestions in addressing RD/RA negotiation time frames: (1) exception reporting and tracking of RD/RA negotiations that exceed 240 days; (2) assuring there is a written policy that establishes a formal process for extending the negotiations beyond 180 days and requires headquarters approval when appropriate; (3) encouraging Regions to consider issuing unilateral orders with effective dates deferred to the end of the new moratorium period when appropriate to help ensure continued progress on long negotiations; and (4) continuing to emphasize early RD starts to assure negotiation time frames do not impact the pace of cleanup. This report represents the opinion of the OIG and does not necessarily represent the final EPA position. In this particular review, the OIG did not measure the offices' performance against the standards established by the National Oil and Hazardous Substances Pollution Contingency Plan (NCP). The results contained in this report are not binding in any enforcement proceeding brought by EPA or the Department of Justice under section 107 of the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) to recover costs incurred not inconsistent with the NCP. We appreciate the opportunity to work with you and your staff in answering some key questions about the RD/RA negotiations process. If you have any questions or need additional information about our work, please contact Fran Tafer, Headquarters Audit Division at (202) 260-2824. 2 EPA OIG Report E1SFG7-11-0024-8400015 ------- PURPOSE. SCOPE AND METHODOLOGY EPA's regional offices and the Office of Enforcement and Compliance Assurance (OECA) have a challenging task integrating competing priorities at the RD/RA negotiation stage. One of the principles underlying EPA's Superfund enforcement program is that those responsible for the release should pay for and correct the damage caused and prevent future problems. At the same time, EPA's Superfund reforms call for fairer treatment of all parties and reduced transaction costs. EPA negotiators must take into account site-specific factors such as the strength of available evidence. They must also work within judicial precedents applicable to their Region. These priorities and other factors affect the extent to which EPA will negotiate with potentially responsible parties (PRPs) to achieve a settlement before using enforcement tools (e.g., unilateral administrative orders, litigation) or Fund-financing to move the cleanup process forward. Purpose We initiated this assignment based on results of past OIG audit work on Superfund administrative improvements and in-depth case studies of three Superfund sites. The objectives were to analyze timeliness of the Agency's RD/RA negotiations; determine the overall trends; and gauge the effect of Superfund enforcement reforms on the timeliness of the negotiations. We conducted the work in cooperation with the Office of Site Remediation Enforcement's (OSRE) Policy and Program Evaluation Division as part of our management advisory services to address questions about timeliness of RD/RA negotiations since 1990. Background Superfund time frames have been a concern to EPA and the Congress. The Agency has initiated several reform efforts over the years in an effort to make the Superfund program work faster and fairer. A growing emphasis on enforcement in the late 1980s culminated in EPA's June 1989 A Management Review of the Superfund Program (90-Day Study) and its recommendation to increase use of enforcement and settlement authorities to induce PRPs to do more cleanups. This approach is sometimes referred to as "enforcement first." The November 1989 Superfund Enforcement Strategy and Implementation Plan (1989 Plan), prepared in response to the House and Senate Appropriations Committees, addressed enforcement issues in more detail. It stated that: "Approximately 80% of negotiations are not fully completed during the moratorium period, although many agreements-in-principle are reached or nearly complete. The Regions were uncertain when unilateral administrative orders (UAOs) should optimally be used." 3 EPA OIG Report E1SFG7-11-0024-8400015 ------- The second broad change affecting Superfund enforcement was announced in June 1993. The Agency issued a series of initiatives to strengthen the Superfund program prior to its reauthorization. Although some Superfund reforms addressed timeliness, the enforcement-related reforms were not meant to address timeliness. One of the goals of these initiatives was "increasing enforcement fairness and reducing transaction costs." As a result, the program focus shifted toward implementing enforcement reforms that would provide for more equitable treatment of PRPs.1 This change in program emphasis was reflected in the Agency's Superfund enforcement priorities for the first time in fiscal year 1993 and was given greater attention in fiscal 1994. Moreover, "enforcement fairness" is a continued priority as indicated by the initiation of additional enforcement reforms in February and October 1995.2 Program management manuals setting out annual priorities added enforcement fairness as a separate program priority in 1993. (Enforcement first remained a priority.) Lengthy enforcement negotiations were identified as a barrier to timely and cost effective cleanups in the 1995 EPA OIG Review of Barriers to Superfund Site Cleanups (Report 6400016, November 29, 1995). During 1997, GAO reported on the broader question of overall time required to clean up Superfund sites, concluding that site cleanup is taking longer than before (Superfund: Times to Complete the Assessment and Cleanup of Hazardous Waste Sites, GAO/RCED-97-20, and Superfund: Duration of the Cleanup Process at Hazardous Waste Sites on the National Priorities List GAO/RCED 97-238R). EPA disagreed with GAO's methodology and conclusions. The Congress held hearings in February 1997 and February 1998 on overall Superfund time frames. Taken together, the law and EPA guidance provide for a 180-day moratorium on EPA remedial action. The purpose of the moratorium is to provide a period for the Agency and PRPs to conduct RD/RA negotiations. The Comprehensive Environmental Response, Compensation and Liability Act (CERCLA), as amended by Superfund Amendments and Reauthorization Act of 1986 (SARA), established, under Section 122 (e), a 120-day moratorium for remedial design/remedial action (RD/RA) negotiations during which the Agency may not start remedial actions. This depends, however, on the PRPs making a good faith offer within 60 days of receiving a special notice letter. EPA policy issued in 1987 allows two thirty-day extensions, increasing the potential moratorium period to 180 days. ^he enforcement reforms in 1993 included greater use of allocation tools (e.g., non-binding allocation and alternative dispute resolution) and de minimis/de micromis settlements as well as evaluating and piloting mixed funding settlement demonstration projects. 2These reforms included expedited settlements, allocation pilots, adopting private party allocations, orphan share compensation, equitable issuance of UAOs, revised de micromis guidance, and site specific special accounts. 4 EPA OIG Report E1SFG7-11-0024-8400015 ------- Subsequent EPA policy and guidance discuss the importance of managing negotiation time frames. The 1987 Interim Guidance: Streamlining the CERCLA Settlement Decision Process stated that, "while the Section 122 provisions related to negotiation moratoria are discretionary, EPA policy will be that those provisions should generally be employed." The 1989 Plan's implementation steps stated that the Regions should be prepared to issue a UAO promptly if (1) a good faith offer is not received within 60 days; or (2) settlement is not reached within 120 days and progress towards settlement is insufficient. Even when the moratorium is extended, the 1989 Plan stated that "Regions should be encouraged to issue unilateral orders with effective dates delayed to the new negotiations deadline." The June 11, 1990 Integrated Timeline for Superfund Site Management states that the "'enforcement first' process is particularly vulnerable to delays if negotiation deadlines are not established or if they are routinely allowed to slip." Scope and Methodology We analyzed one part of the overall Superfund process, RD/RA negotiations. The scope of our analysis was national, focusing primarily on the fiscal 1990-1996 period. In addition to analyzing national data on RD/RA negotiation time frames, we met with Headquarters staff and Superfund remedial project managers (RPMs) and attorneys in Region 3 and Region 5 to discuss results of the analysis and factors affecting the length of RD/RA negotiations, including Superfund reforms. We analyzed National Priorities List sites, excluding state-lead and Federal facility sites. To analyze the timeliness of the Agency's RD/RA negotiations, we used several methods to identify overall trends. We concentrated on changes in EPA policy and guidance because it should affect how RD/RA negotiations are conducted and can be controlled by the Agency. We used these analytical methods to address the second key objective of gauging the effect of Superfund enforcement reforms on the timeliness of the negotiations. We discussed our results with OSRE periodically and used their suggestions for additional analyses. As we looked for the effect of the reforms, we also saw the earlier effect of the 90-Day Study and related policy changes. We calculated fiscal year averages for: (1) negotiations started in fiscal 1990- 1996 and (2) negotiations completed in fiscal 1990-1996. Although these two calculations include many of the same negotiations, they produced very different results, as seen below in this report. We also analyzed the percentage of negotiations completed within the moratorium period and beyond and calculated median values. 5 EPA OIG Report E1SFG7-11-0024-8400015 ------- To further analyze how the Superfund enforcement reforms may have affected the duration of RD/RA negotiations, we chose to examine durations before and after that change. First, we excluded the initial year in which the reforms were announced (fiscal 1993). Then we calculated (1) the average length of negotiations completed in fiscal 1990-1992, before reforms began; and (2) the average length of negotiations started in 1994-1996, after reforms were underway. The negotiations completed in 1990-1992 would not have been influenced by the reforms, but those started in 1994- 1996 may have been affected. We used the same methodology to analyze the effect of the 90-Day Study. For 12 negotiations started within our time period that were still ongoing when we concluded our work, we assigned a temporary ending date of January 31, 1998. This allowed us to include these ongoing negotiations in our calculations based on start dates. Without including them, we believed the averages calculated by starting date would have been overly optimistic because they would have excluded the on-going negotiations. Please see Appendix 2 for a more detailed discussion of scope and methodology. This work was completed in accordance with generally accepted government auditing standards except for the fourth field work standard on management controls. We did not formally assess the validity and reliability of CERCLIS data or controls. 6 EPA OIG Report E1SFG7-11-0024-8400015 ------- RESULTS AND MATTERS FOR CONSIDERATION Average RD/RA Negotiation Time Frames Have Changed Over Time We found different patterns when calculating the average duration of RD/RA negotiations by fiscal year, depending on whether we used negotiation start dates or completion dates. The trend shown by the completion date analysis seems to parallel the broad policy changes discussed above; however, the trend based on starting dates is less clear. (See Figure 1.) Coincidental^, the actual number of negotiations started and completed declined from fiscal 1991 through fiscal 1996. (See Figure 2). Figure 1 Average Length of RD/RA Negotiations: 1988-1996 420 360 300 « 240 Q 180 120 1988 1989 1990 1991 1992 1993 1994 1995 1996 Fiscal Year Based on Date Started Based on Date Completed Fiscal Year Average 1988 1989 1990 1991 1992 1993 1994 1995 1996 Based on Date Started 219 312 183 229 266 276 254 284 267 Based on Date Completed 251 227 326 190 214 203 326 372 367 7 EPA OIG Report E1SFG7-11-0024-8400015 ------- Figure 2 Number of RD/RA Negotiations: 1988-1996 160 140 120 100 60 40 20 1988 1989 1990 1991 1992 1993 1994 1995 1996 Fiscal Year «— # Started —¦— #Completed Fiscal Year 1988 1989 1990 1991 1992 1993 1994 1995 1996 # Started 117 108 98 154 104 100 68 66 43 # Completed 98 108 117 134 118 84 90 68 58 As can be seen in Figure 1, fiscal year averages based on completion dates form a dramatic U-shaped curve between 1990 and 1994. Superfund enforcement policy changes in fiscal 1989 and 1990 appear to have set the stage for a 42 percent drop (136 days) in average RD/RA negotiation time frames-from 326 days in fiscal 1990 to 190 days in 1991. The drop occurred about one fiscal year after EPA's "enforcement first" emphasis was highlighted in the 90-Day Study and the 1989 Plan. Similarly, negotiation times rose 61 percent (123 days), from 203 days in 1993 to 326 days in fiscal 1994, the year after enforcement fairness was added as a program priority. (The higher numbers in 1990 and 1994-1996 may also be due to the Agency completing some negotiations that had been in process for a long time. We discuss this possibility below in sections on median values and proportion completed within the moratorium period.) The picture is less obvious when RD/RA negotiating time is calculated using the dates negotiations were started. For example, after the 90-Day Study was issued, average time declined from 312 days in fiscal 1989 to 183 days in fiscal 1990, a 41 percent (129 day) reduction. No similarly dramatic change occurred after the Superfund enforcement reforms were introduced. In fact, average time for negotiations 8 EPA OIG Report E1SFG7-11-0024-8400015 ------- started each year rose steadily after fiscal 1990, except for small drops in 1994 and 1996. Negotiations started in fiscal 1993 and 1995, when the Superfund reforms were issued, were longer than other years, but the variation between 1992 and 1996 is relatively small (no more than 30 days). When graphed, the starting date averages form a pattern from fiscal 1988-1992 that is roughly similar to the pattern formed by the completion date averages for fiscal 1989-1993. To see how closely the starting date averages seemed to predict the following year's completion date averages, we took our original graph (Figure 1) and moved the starting date averages one year forward. (See Figure 1a.) Figure 1a Comparison of Average Time Frames 420 360 O 180 120 60 1989 1990 1991 1992 1993 1994 1995 1996 Completion Date Fiscal Year —•— Starts —¦— Completions Note: Fiscal year averages based on date started were moved forward one year in this table and chart for comparison purposes. Fiscal Year Average 1989 1990 1991 1992 1993 1994 1995 1996 Based on Date Started 219 312 183 229 266 276 254 284 Based on Date Completed 227 326 190 214 203 326 372 367 The close relationship in the earlier years may simply reflect that RD/RA negotiations started in one fiscal year were being completed in the following year. If so, then Figure 1a suggests that something occurred to change this pattern beginning with negotiations started in fiscal 1992 or completed in fiscal 1993. 9 EPA OIG Report E1SFG7-11-0024-8400015 ------- Medians Show Increase in 1995 The median (or middle value) helps show what happened to RD/RA negotiation time frames because it is affected less by particularly long or short negotiations than the average (or mean). Median number of days for each fiscal year showed only modest changes until fiscal 1995 when the median for RD/RA negotiations started that year increased 49 percent (80 days) over fiscal 1994. Similarly, the median for negotiations completed that year increased 38 percent (71 days) over fiscal 1994 (from 188 to 259 days). Interestingly, the medians for both negotiations started and completed are similar for fiscal 1990-1996. They show little variation from fiscal 1990- 1993, unlike the average based on completion dates which shows a dramatic decline from fiscal 1990 to 1991. (Compare Figures 1 and 3.) This suggests that some of the negotiations completed in fiscal 1990 may have been particularly long, raising the average but not the median. The significant increase in the median for RD/RA negotiations started or completed in fiscal 1995 and fiscal 1996 suggests, along with higher averages, that negotiation time frames headed upwards in those years. Figure 3 Median Length of RD/RA Negotiations 420 360 300 «240 Q 180 120 60 1990 1991 1992 1993 1994 1995 1996 Fiscal Year Based on Date Started Based on Date Completed Fiscal Year Medians 1990 1991 1992 1993 1994 1995 1996 Based on Date Started 167 173 183 187 164 244 217 Based on Date Completed 184 170 178 181 188 259 244 Smaller Proportion Completed Within the Moratorium The percentage of RD/RA negotiations completed within the 180-day moratorium period has declined to under 40 percent in fiscal 1995-1996. Conversely, the percentage of long negotiations, (those over 240 days), was higher in those years than the preceding five years. OSRE believes that the 240 day mark is significant, because 10 EPA OIG Report E1SFG7-11-0024-8400015 ------- it takes into account the statutory 120-day moratorium, the two 30-day extensions allowed by EPA guidance, and an estimated additional 60 days to prepare a UAO (assuming that the need for a UAO would become clear only after a full 180 day negotiation period). This pattern is consistent when analyzing the data based on negotiation starting dates and completion dates (see Figures 4 and 5 and Appendix 3). A greater percentage of negotiations started in fiscal 1995-1996 took longer than previous years. Similarly, a greater percentage of negotiations completed in fiscal 1995-1996 took longer than previous years. 3 Figure 4 Percentage Breakout by Length—Based on Year Started 1996 re CD >- re o (A 995 994 993 992 991 990 10 20 30 40 50 60 70 80 90 100 Percentage 0-180 Days 240-365 Days 2-3 Years 4-5 Years 180-240 Days 1-2 Years 3-4 Years Over 5 Years re u >- re o > Figure 5 Percentage Breakout by Length—Based on Year Completed 996 -t 995 994 993 992 991 990 10 20 30 40 50 60 70 80 90 100 Percentage 0-180 Days 240-365 Days 2-3 Years 4-5 Years 180-240 Days 1-2 Years 3-4 Years Over 5 Years 3 For data underlying Figures 4 and 5, see Appendix 3. 11 EPA OIG Report E1SFG7-11-0024-8400015 ------- In fact, over 16 percent of negotiations completed in fiscal 1995 and 1996 took over 2 years. OSRE officials believe that the Superfund enforcement reforms may have been a factor in bringing some of the older negotiations to closure by addressing issues of concern to PRPs. Policy Appears to Affect Time Frames We found that policy changes appeared to affect average RD/RA negotiation time frames. To better evaluate the effect of policy changes, we excluded the fiscal year in which the Superfund reforms began (1993) and the year the "enforcement first" emphasis began (1989). Then we calculated average durations for negotiations completed before each change and compared them to averages for negotiations started after the change. The results show that negotiating time was lower after the "enforcement first" emphasis was introduced and higher after reforms began. Specifically, the average RD/RA negotiation time frame declined from 272 days for the three years preceding the 90-Day Study to a 227 day average for the three years after it was introduced-this represents a 17 percent decrease (45 days). Conversely, the average for fiscal 1990- 1992 was 241 days, compared to 268 days for fiscal 1994-1996, the three years following the introduction of reforms in 1993—this represents an 11 percent rise (27 days). (See Figures 6 and 7.) Figure 6 Figure 7 Time Frames Before & After 90-Day Study Time Frames Before & After Start of Superfund Reforms ~ FY 86-FY 88 ~ FY 90 - FY 92 ~ FY 90 - FY 92 ~ FY 94 - FY 96 Although the effect of policy changes is apparent at the national level, the report sections that follow point out that regional and site-specific factors come into play as well. 12 EPA OIG Report E1SFG7-11-0024-8400015 ------- Regional Emphasis Mav Influence RD/RA Negotiation Time Frames Different regional management approaches appear to have influenced RD/RA negotiations at the two regional offices we visited. Region 3 and Region 5 together account for about 1/3 of the 669 RD/RA negotiations completed nationally from fiscal years 1990-1996. Region 3 completed 108 negotiations and Region 5 completed 120 negotiations during this period. Figures 8 through 11 show the two Regions' number of negotiations and average time to complete the negotiations. Figure 8 Region 3 Average Times >.420 Figure 9 Region 3 Number of RD/RA Negotiations 1990 1991 1992 1993 1994 1995 1996 Fiscal Year 1990 1991 1992 1993 1994 1995 1996 Fiscal Year Starts Completions Starts Completions Fiscal Year Averages 1990 1991 1992 1993 1994 1995 1996 Based on Date Started 198 152 335 445 472 416 302 Based on Date Completed 312 123 173 246 406 408 738 Fiscal Year 1990 1991 1992 1993 1994 1995 1996 # Started 10 24 25 17 9 8 8 # Completed 21 17 23 14 11 13 9 Figure 10 Region 5 RD/RA Negotiation Time Frames 780 720 660 600 540 480 >,420 ^360 300 240 180 120 60 0 Figure 11 Region 5 Number of RD/RA Negotiations 1990 1991 1992 1993 1994 1995 1996 Fiscal Year 1990 1991 1992 1993 1994 1995 1996 Fiscal Year Starts Completions Starts Completions Fiscal Year Average 1990 1991 1992 1993 1994 1995 1996 Based on Date Started 189 199 192 215 135 241 137 Based on Date Completed 215 205 202 125 198 172 186 Fiscal Year 1990 1991 1992 1993 1994 1995 1996 # Started 16 31 20 21 13 11 4 # Completed 17 25 28 19 16 6 9 13 EPA OIG Report E1SFG7-11-0024-8400015 ------- Region 3 fiscal year averages were higher than the national averages in fiscal 1993- 1996, although they were lower for negotiations completed in fiscal 1990-1992. (See Figures 1 and 8.) Region 5 RD/RA negotiations were generally shorter than the national average. (See Figure 10.) Region 3 focuses on recovering past costs which is achieved through settlements with potentially responsible parties (PRPs). They emphasize a willingness to listen to PRP issues. In addition, we were informed that Region 3 works with somewhat different judicial precedents than Region 5. Region 5 told us they place greater emphasis on negotiation deadlines in order to get cleanups started. Region 3 informed us that they instituted a process during fiscal 1997 requiring written approval to extend negotiations beyond 180 days. If an extension is not approved, then negotiations would be terminated. Based on our discussions with Agency officials, we believe there are trade-offs with each approach. Emphasizing settlements may lengthen the negotiations. We believe that longer negotiations would delay the point at which cleanup actions begin. However, analyzing duration through RD start was beyond the scope of this review and officials pointed out that some PRPs begin work on the RD before negotiations are completed. Even if negotiations take longer and the start of cleanup is delayed, the PRPs may be more cooperative overall and the cleanup process may proceed more smoothly and quickly once it begins. On the other hand, focusing on negotiation time frames and issuing UAOs when the moratorium period expires could result in early initiation of the cleanup process. If applied consistently overtime, this approach might cause PRPs to settle more quickly if they want to avoid UAOs. However, UAOs generally require the Region to follow up with a separate cost recovery case. Settlements usually include cost recovery. OSRE noted that guidance issued in March 1990 indicates an Agency preference for settlements. Officials in both regional offices expressed a preference for settlements. Case Specific Factors Affect Time Frames Although management direction at the national and regional levels appear to affect RD/RA negotiation average time frames, there are a number of case-specific factors that can shorten or lengthen RD/RA negotiations. Some of these factors are clearly not within EPA's control. Based on our interviews with regional officials, we found that the sites with relatively short RD/RA negotiation time frames were not necessarily the result of "best practices." Generally, these were sites where the Agency was confident there was little or no chance of reaching a settlement and proceeded to issue a UAO. The sites with longer RD/RA negotiation time frames were affected by a number of factors, many of which were case-specific and not controllable by the Agency. For example, at one site the RD/RA negotiations took over 1000 days due to ongoing 14 EPA OIG Report E1SFG7-11-0024-8400015 ------- litigation in Federal district and appeals courts. At another site, PRPs could not reach an agreement about their cost share, which accounted for more than two years of negotiation time. In addition to asking what caused delays at specific sites, we asked Agency officials to provide factors they believe typically delay negotiations. They mentioned a number of factors, including cost of the remedy, identifying additional PRPs, Natural Resource Trustee issues and ongoing litigation. See Appendix 1 for specifics. CONCLUSIONS Many factors can affect how long individual RD/RA negotiations last. Some of these factors may not be within EPA's control. However, the change in negotiation time frames between fiscal 1990 and fiscal 1996 appears to reflect changes in management direction and approach during that time period. An early emphasis on statutory and policy time frames has competed more recently with the Superfund reforms' emphasis on enforcement fairness. Consequently, time frames for this one piece of the Superfund "pipeline" have lengthened. Longer RD/RA negotiating time could mean a delay in getting cleanup activities started, depending on when the RD is started. On the other hand, enforcement fairness might result in less contentious and quicker cleanup and cost recovery. MATTERS FOR CONSIDERATION We have not included formal recommendations because this report is designed to provide information as part of the OIG's Advisory Services. The Director, OSRE may want to consider the following suggestions in addressing RD/RA negotiation time frames: (1) exception reporting and tracking of RD/RA negotiations that exceed 240 days; (2) assuring there is a written policy that establishes a formal process for extending the negotiations beyond 180 days and requires headquarters approval when appropriate; (3) encouraging Regions to consider issuing unilateral orders with effective dates deferred to the end of the new moratorium period when appropriate to help ensure continued progress on long negotiations; and (4) continuing to emphasize early RD starts to assure negotiation time frames do not impact the pace of cleanup. OSRE provided us Agency comments during a teleconference and through an edited version of the draft report. They agreed overall with the report, conclusions and matters for consideration. They also suggested we change wording to clarify points, correct a definition, and expand the matters for consideration to include emphasizing early RD starts. We also received informal comments from an official in OSWER and officials we met with in Regions 3 and 5. We adopted their suggestions in most cases. We added more categories to Figures 4 and 5 to show the percentages of older negotiations started and completed each fiscal year. We also added wording to reflect 15 EPA OIG Report E1SFG7-11-0024-8400015 ------- OSRE opinions that (1) the Superfund enforcement reforms may have helped in closing out some of the older negotiations; and, (2) longer negotiations would not necessarily delay the start of cleanup, depending on when the RD is started. 16 EPA OIG Report E1SFG7-11-0024-8400015 ------- Appendix 1 ADDITIONAL FACTORS AFFECTING RD/RA NEGOTIATION TIME FRAMES In addition to asking what caused delays at specific sites, we asked Agency officials to provide factors they believe typically delay negotiations. The following table lists the factors regional officials in both Regions 3 and 5 cited as causing delay in the RD/RA negotiation process. Factor Explanation Cost of the Remedy PRPs faced with remedies that are relatively high in cost will sometimes request additional time and information before reaching an agreement. Identification of Additional PRPs Additional PRPs were discovered and brought in to the process after negotiations had been initiated. Natural Resource Trustee (NRT) Involvement NRTs became involved late in the process; made unrealistic cleanup demands. De minimis settlements/ Number of PRPs The more parties involved, the longer it takes to negotiate. Ongoing Litigation Issues being litigated that affected the outcome of negotiations. PRP Experience If a PRP has not had prior experience with Superfund, then negotiations tend to take longer. PRP Disagreed with the Remedy PRPs wanted changes to and/or different remedies. PRP Cost Allocation PRPs inability to agree among themselves on the division of costs. State Issues/Involvement Addition of another party to the process; inability to pay cost share; late involvement; requests for stricter cleanup standards than those proposed by the Agency. Viability of PRPs The financial status of the PRPs and their ability to fund a Superfund cleanup. Weak Evidence (Litigation Risk) Weak evidence increases litigation risk, which weakens the Agency's negotiating position. Other factors mentioned in a Region or by OSRE included: PRP view of liability; 17 EPA OIG Report E1SFG7-11-0024-8400015 ------- Appendix 1 orphan share; type of site (i.e., landfill); cost recovery issues; Congressional interest/politics; litigious approach and personality of opposing counsel; involvement of a Federal PRP; initial use of a reform; PRP delay tactics/corporate philosophy; lack of incentive for parties to settle; lack of continuity of participation (turnover of people involved); local government involved; community objections to the remedy; remedy was reconsidered; model CD; no fund-lead dollars available for the site; and, ability-to-pay determinations. 18 EPA OIG Report E1SFG7-11-0024-8400015 ------- Appendix 2 SCOPE AND METHODOLOGY Scope The scope of this analysis includes all RD/RA negotiations for National Priorities List (NPL) sites completed by EPA, excluding state-lead sites and federal facilities. We limited our work primarily to fiscal 1990-1996. We started with 1990 because the Superfund program underwent a change in enforcement emphasis about that time, starting with the 90-Day Study, "A Management Review of the Superfund Program." That report called on the Agency to "emphasize enforcement to induce private party cleanup." It stated that "EPA will use the fact and threat of enforcement, encompassing a broad range of administrative and legal tools, to increase the proportion of cleanups undertaken by private parties." Based on our past audit work and discussions with OSRE, we decided to use 1990 as our starting point because of this increased emphasis on "enforcement first." Our fieldwork was conducted from July 1997 to January 1998. Methodology Our methodology included primarily (1) analysis of CERCLIS data; (2) reviews of relevant law, regulations, policy and guidance; and, (3) interviews of Agency staff at headquarters, Region 3, and Region 5. We also met with PRP representatives to get a better appreciation of their view of the RD/RA negotiation process. Based on our analysis of CERCLIS data and discussions with EPA staff, we decided to visit Region 3 and Region 5. They were among the top three Regions in numbers of negotiations and they showed significantly different results in RD/RA negotiation times. In Region 3 and Region 5, we discussed with attorneys and RPMs some specific negotiations that took comparatively long and some that took a comparatively short time to identify whether lessons learned could be generalized to be used as best practices to share throughout the Agency. We also discussed the overall trends for each Region and the effect of the Superfund reforms on the RD/RA negotiation process. CERCLIS. We used CERCLIS report ENFR59, "Negotiations Master Report" data in electronic form covering data through June 1997. We did not include fiscal year 1997 data in our analysis because the CERCLIS data for fiscal year 1997 was not complete. In addition, we did not test CERCLIS data reliability and accuracy; however, we did select a standard CERCLIS report for our analysis, which Agency officials told us had better controls than an ad hoc report. In addition, OSRE officials informed us that regional staff checked and corrected data as a result of our review. Calculating Length of Negotiations. We calculated the number of calendar days between the actual RD/RA negotiation start date and the actual RD/RA negotiation 19 EPA OIG Report E1SFG7-11-0024-8400015 ------- Appendix 2 completion date as indicated in the CERCLIS ENFR59 report. First, we calculated fiscal year averages for: (1) negotiations started in fiscal 1990-1996 and (2) negotiations completed in fiscal 1990-1996. Although these two methods include many of the same negotiations, they produced very different results. (See Figure 1.) To assess the impact of two major policy changes, we calculated how long negotiations took during the years immediately preceding and following the change. (See Figures 6 and 7.) We also analyzed the percentage of negotiations completed within the moratorium period and beyond. (See Figures 4 and 5.) In addition, we calculated median values. (See Figure 3.) For 12 negotiations started within our time period that were still ongoing when we concluded our work, we assigned a temporary ending date of January 31, 1998. This allowed us to include these ongoing negotiations in our calculations based on start dates. Without including them, we believed the averages calculated by starting date would have been overly optimistic because they would have excluded the on- going negotiations. In addition, we tested the effect of outliers (negotiations under 10 days and over 1460 days or 4 years) on the overall trends. We found that, although the average times varied somewhat, the overall trends remained the same. Therefore, we included all negotiations in our calculations. Figures 12 and 13 below show the trend excluding the outliers. Figure 12 RD/RA Negotiation Time Frames - Excluding Outliers 420 360 300 o 180 120 60 1988 1989 1990 1991 1992 1993 1994 1995 1996 Fiscal Year — Starts —¦— Completions Fiscal Year Average 1988 1989 1990 1991 1992 1993 1994 1995 1996 Based on Date Started 233 286 190 224 276 280 274 303 302 Based on Date Completed 254 219 338 212 232 221 290 371 343 20 EPA OIG Report E1SFG7-11-0024-8400015 ------- Appendix 2 Figure 13 Number of RD/RA Negotiations - Excluding Outliers 160 140 120 100 60 40 20 1988 1989 1990 1991 1992 1993 1994 1995 1996 Fiscal Year •— # Started —¦— # Completed Fiscal Year 1988 1989 1990 1991 1992 1993 1994 1995 1996 # Started 110 104 94 138 94 92 63 62 38 # Completed 90 105 113 120 109 77 84 62 51 Communication. As part of our methodology, we worked closely with OSRE evaluation staff, discussing our approach and methodology to ensure what we did made sense to them and met their needs for information about the RD/RA negotiation process. This two-way communication allowed us to benefit from their suggestions on how to analyze the issue. This was in accordance with OIG policy/guidance on providing assistance to the Agency in improving program management. Standards. This work was completed in accordance with generally accepted government auditing standards except for the fourth field work standard on management controls. We did not formally assess the validity and reliability of controls over CERCLIS data. Limitations. This report represents the opinion of the OIG and does not necessarily represent the final EPA position. In this particular review, the OIG did not measure the offices' performance against the standards established by the National Contingency Plan (NCP). The results contained in this report are not binding in any enforcement proceeding brought by EPA or the Department of Justice under section 107 of the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) to recover costs incurred not inconsistent with the NCP. 21 EPA OIG Report E1SFG7-11-0024-8400015 ------- Appendix 3 TABLES FOR FIGURES 4 AND 5 Percentage Breakout by Length-Based on Year Started Fiscal Year in Which Negotiations Were Started % That Took 1990 1991 1992 1993 1994 1995 1996 0-180 Days 53% 53 47 47 56 36 35 180-240 Days 21 22 15 20 15 14 21 240-365 Days 23 16 18 13 10 30 12 1 -2 Years 2 5 13 10 10 12 32 2-3 Years 0 2 4 8 7 5 0 3-4 Years 1 1 2 1 2 3 0 4-5 Years 0 0 1 1 0 0 0 Over 5 Years 0 1 0 0 0 0 0 Percentage Breakout by Length-Based on Year Completed Fiscal Year in Which Negotiations Were Completed % That Took 1990 1991 1992 1993 1994 1995 1996 0-180 Days 49 52 52 49 47 37 31 180-240 Days 9 25 16 21 20 12 19 240-365 Days 10 17 23 19 9 21 22 1 -2 Years 19 4 7 10 14 13 12 2-3 Years 8 2 1 0 6 12 10 3-4 Years 5 0 1 1 2 4 2 4-5 Years 0 0 0 0 1 0 2 Over 5 Years 0 0 0 0 1 1 2 22 EPA OIG Report E1SFG7-11-0024-8400015 ------- Appendix 4 DISTRIBUTION LIST Inspector General (2410) Assistant Administrator, OECA (2201 A) Principal Deputy Assistant Administrator, OECA (2201 A) Director, Regional Support Division, OSRE, OECA (2272 A) Audit Follow-up Coordinator, OECA (2201 A) Audit Follow-up Coordinator, OSWER (5101) Agency Follow-up Official (2710), Attn: Director, Resource Management Division, OARM Agency Follow-up Coordinator (2724) Comptroller (2731) Director, Grants Administration Division (3903 F) Associate Administrator for Congressional and Legislative Affairs (1301) Associate Administrator for Communications, Education, and Public Affairs (1701) Administrator, Region 3 Director, Hazardous Waste Management Division, Region 3 Audit Follow-up Coordinator, Region 3 Office of External Affairs, Region 3 Administrator, Region 5 Director, Superfund Division, Region 5 Audit Follow-up Coordinator, Region 5 Office of Public Affairs, Region 5 23 EPA OIG Report E1SFG7-11-0024-8400015 ------- |