Superfund

Since the Superfund program began in 1980, the Environmental Protection Agency (EPA) has spent about $17.7 billion to clean up hazardous waste sites. EPA oversees the work, but private contractors do the actual cleanups--known as remedial actions--of Superfund sites. GAO has issued a series of reports that describe the portion of EPA's funds spent on remedial actions by contractors, as opposed to other activities, such as studying conditions at sites, designing cleanup remedies, and travel. (See GAO/RCED-97-211, Sept. 1997, GAO/RCED-98-221, Aug. 1998, and GAO/RCED-99-139, May 1999.) This report examines the costs at the following five Superfund sites: the Raymark site in Stratford, Connecticut; the Sharon Steel site in Midvale, Utah; the United Creosoting site in Conroe, Texas; the NL Industries site in Granite City, Illinois; and the Newmark site in San Bernadino, California. For each site, GAO discusses (1) what portion of the total funds EPA spent on each site was used to pay contractors for remedial actions as opposed to other activities and how the contractors spent the money and (2) whether the actual costs for remedial actions differed from the estimated costs and why.

GAO noted that: (1) most of the funds that EPA spent as of May 1999 at each of the five Superfund sites went to contractors for implementing remedial actions at the sites; (2) these contractors include the prime contractors that generally manage cleanups and the subcontractors that do the physical cleanup work; (3) the cost for these prime contractors and subcontractors ranged from 53 percent to 86 percent of EPA's total spending at the sites GAO visited; (4) the costs for site studies and remedial designs were the second largest portion of the total costs at three of the five sites, ranging from about 7 percent to about 33 percent; (5) at three of the five sites, the prime contractors spent the bulk of the funds they received from EPA either on subcontractors that performed the physical cleanup or on supplies, equipment, and in-house labor associated with the sites' cleanup; (6) at two other sites, details on contractors' spending were not available because the sites were cleaned up under fixed-price contracts, which do not require contractors to report how they spend funds; (7) work at two of the five sites was still in progress at the time of GAO's review, so both total expenditures and expenditures on remedial actions will increase over time; (8) at the five sites GAO visited, for the remedial cleanup activities that had occurred as of May 1999, the relationship between the actual costs for these activities and the estimated costs varied: (a) at three of the sites, the actual costs exceeded the estimated costs; (b) at one site, the actual costs were lower than the estimated costs; and (c) at the remaining site, a meaningful comparison was not possible because the full scope of the work and its costs were not known at the start of cleanup; (9) at the three sites where actual cost exceeded estimated costs, the total cleanup costs ranged from less than 1 percent to 15 percent higher than estimated; (10) these increases were attributable to various factors, including higher-than-anticipated quantities of materials and supplies; and (11) costs were lower than estimated at one site because, although EPA spent about $30 million on an innovative cleanup method that failed, the actual amount of contaminated soil the remedy had to address turned out to be only about one-forth of the amount initially estimated.