This is the accessible text file for GAO report number GAO-07-592 entitled 'National Animal Identification System: USDA Needs to Resolve Key Implementation Issues to Achieve Rapid and Effective Disease Traceback' which was released on August 3, 2007. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to the Chairman, Committee on Agriculture, Nutrition, and Forestry, U.S. Senate: United States Government Accountability Office: GAO: July 2007: National Animal Identification System: USDA Needs to Resolve Several Key Implementation Issues to Achieve Rapid and Effective Disease Traceback: GAO-07-592: GAO Highlights: Highlights of GAO-07-592, a report to the Chairman, Committee on Agriculture, Nutrition, and Forestry, U.S. Senate Why GAO Did This Study: Livestock production contributed nearly $123 billion to the U.S. economy in 2006. In response to concerns about animal disease outbreaks, the U.S. Department of Agriculture (USDA) announced in December 2003 that it would implement a nationwide program—later named the National Animal Identification System (NAIS)—to help producers and animal health officials respond quickly and effectively to animal disease events in the United States. In this context, GAO determined (1) how effectively USDA is implementing NAIS and, specifically, the key issues identified by livestock industry groups, market operators, state officials, and others; (2) how USDA has distributed cooperative agreement funds to help states and industry prepare for NAIS and evaluated the agreements’ results; and (3) what USDA and others estimate are the costs for USDA, states, and industry to implement NAIS. In conducting its work, GAO reviewed USDA documents; interviewed agency, industry, and state officials; and consulted 32 animal identification (ID) experts. What GAO Found: In implementing the NAIS program, USDA has taken some steps to address issues identified by livestock industry groups, market operators, state animal health officials, and others. Nonetheless, the agency has not effectively addressed several issues that, if left unresolved, could undermine the program’s ability to achieve the goal of rapid and effective animal disease traceback. Specifically, USDA’s decision to implement NAIS as a voluntary program may affect the agency’s ability to attract the necessary levels of participation. However, some industry groups believe that NAIS could succeed as a voluntary program, or that USDA needs to first resolve several issues before making participation mandatory. Agency officials are analyzing what participation levels are necessary to meet the program’s goal and may introduce benchmarks to measure progress. In addition, several key problems hinder USDA’s ability to implement NAIS effectively: • USDA has not prioritized the implementation of NAIS by species or other criteria. Instead, the agency is implementing NAIS for numerous species simultaneously, causing federal, state, and industry resources to be allocated widely, rather than being focused on the species of greatest concern. * USDA has not developed a plan to integrate NAIS with preexisting USDA and state animal ID requirements. As a result, producers are generally discouraged from investing in new ID devices for NAIS. * USDA has not established a robust process for selecting, standardizing, and testing animal ID and tracking technologies. * USDA does not clearly define the time frame for rapid traceback, possibly slowing response and causing greater economic losses. * USDA does not require potentially critical information to be recorded, such as species or age, in the NAIS databases. USDA has awarded $35 million in NAIS cooperative agreements from fiscal years 2004 through 2006 to 49 states, 29 tribes, and 2 territories to help identify effective approaches to register premises and identify and track animals. However, USDA has not consistently monitored or formally evaluated the results of cooperative agreements or consistently shared the results with states, industry groups, and other stakeholders. As a result, USDA cannot be assured that the agreements’ intended outcomes have been achieved and, furthermore, that lessons learned and best practices are used to inform the program’s progress. No comprehensive cost estimate or cost-benefit analysis for the implementation and maintenance of NAIS currently exists. As a result, it is not known how much is required in federal, state, and industry resources to achieve rapid and effective traceback, or whether the potential benefits of the program outweigh the costs. Industry groups and state officials say the cost of implementing NAIS is one of their biggest concerns. USDA plans to hire a contractor to conduct a cost- benefit analysis, in part to more precisely forecast the economic effects of NAIS. What GAO Recommends: GAO made several recommendations to help USDA achieve the program’s goal of rapid and effective animal disease traceback. In commenting on a draft of this report, USDA generally agreed with the recommendations. [Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-07-592]. To view the full product, including the scope and methodology, click on the link above. For more information, contact Lisa Shames at (202) 512- 3841 or ShamesL@gao.gov. [End of section] Contents: Letter: Results in Brief: Background: Key NAIS Implementation Issues Are Unresolved and Could Undermine the Program's Goal of Rapid and Effective Traceback: USDA Has Awarded NAIS Cooperative Agreements to Identify Effective Implementation Approaches but Has Not Formally Evaluated Agreements' Results: Total NAIS Program Costs Have Not Been Determined, but USDA Recently Announced Plans to Develop a Cost-benefit Analysis: Conclusions: Recommendations for Executive Action: Agency Comments and Our Evaluation: Appendix I: Objectives, Scope, and Methodology: Appendix II: Select Domestic and Foreign Animal Diseases of Concern Identified by USDA: Appendix III: Members of GAO's Expert Panel on NAIS: Expert Panel Members: Appendix IV: GAO Expert Panel Questions and Responses on NAIS: Expert Panel: USDA's Implementation of the National Animal Identification System: Appendix V: Select International Animal Identification and Tracking Programs: Appendix VI: Information on NAIS Cooperative Agreements: Appendix VII: Comments from the U.S. Department of Agriculture: GAO Comments: Appendix VIII: GAO Contact and Staff Acknowledgments: Tables: Table 1: USDA NAIS Budget Data, Fiscal Years 2004 (CCC Funds) through 2006: Table 2: USDA Funding for NAIS Cooperative Agreements to States, Territories, and Tribes, Fiscal Years 2004 through 2006: Table 3: Species Covered by USDA NAIS Cooperative Agreement Field Trials, Fiscal Years 2004 and 2005: Table 4: NAIS Premises Registration Statistics, by State, as of June 4, 2007: Table 5: USDA's Criteria for Distributing NAIS Cooperative Agreement Funding in Fiscal Year 2007: Figures: Figure 1: A Calf Identified with Both a Visual and an Electronic Tag: Figure 2: Number of NAIS Premises Registered, 2005 to Present: Abbreviations: APHIS: Animal and Plant Health Inspection Service: BSE: bovine spongiform encephalopathy: CCC: Commodity Credit Corporation: EU: European Union: FMD: foot-and-mouth disease: ID: identification: NAIS: National Animal Identification System: OMB: Office of Management and Budget: RFID: radio frequency identification: TB: tuberculosis: USDA: U.S. Department of Agriculture: United States Government Accountability Office: Washington, DC 20548: July 6, 2007: The Honorable Tom Harkin: Chairman: Committee on Agriculture, Nutrition, and Forestry: United States Senate: Dear Mr. Chairman: U.S. agriculture provides an abundant supply of food and other products for Americans and others around the world, annually generating more than $1 trillion in economic activity, including more than $68 billion in exports in 2006. Because of the economic importance of the agriculture sector and the risks to public health as well as the economy, we have designated the federal oversight of food safety as a high-risk area.[Footnote 1] Within the broader sector, livestock production contributed nearly $123 billion to the U.S. economy in 2006, including $13.4 billion in livestock, poultry, and dairy exports.[Footnote 2] One way to protect the health of livestock animals--which are critically important to the integrity and safety of the nation's food supply, the well-being of Americans, and the U.S. economy--is through a national animal identification (ID) system to trace back and contain diseases that spread rapidly. Our recent work has described animal diseases and their economic and, in some cases, human health consequences. For example, a highly pathogenic strain of avian influenza has spread to nearly 60 countries over the past few years, resulting in the death and destruction of millions of wild and domestic birds and infecting almost 300 humans, more than one-half of whom have died--creating serious concerns that the virus could reach North America at any time.[Footnote 3] In addition, the first known U.S. case discovered in December 2003 of one cow infected with bovine spongiform encephalopathy (BSE) caused the U.S. beef industry to lose more than 80 percent of its export trade, or an estimated $2 billion, between January and September 2004.[Footnote 4] Commonly known as mad cow disease, BSE has been linked by scientists to a fatal neurological disease in humans known as variant Creutzfeldt-Jacob disease. Another disease of particular concern is foot-and-mouth disease (FMD), a highly contagious livestock disease that does not typically affect humans and last occurred in the United States in 1929. According to several estimates, the direct costs of controlling and eradicating a U.S. outbreak of FMD could range up to $27 billion in current dollars.[Footnote 5] In response to concerns about such outbreaks occurring in the United States and in recognition that speed and accuracy are critical factors in controlling a disease, the U.S. Department of Agriculture (USDA) announced in December 2003 that it would lead the design and implementation of a nationwide program--later named the National Animal Identification System (NAIS)--to enable USDA, states, and industry to quickly and efficiently locate all infected and potentially exposed animals and premises that have had contact with a foreign or domestic disease of concern. USDA recognized that a fully functional animal tracking system will keep the United States competitive in international markets, can help reassure foreign consumers about the health of U.S. livestock, and may satisfy other countries' import requirements. Internationally, some of the United States' major trading partners--such as the European Union (EU), Japan, and Canada--already have mandatory national animal ID programs in place for certain species. The Animal Health Protection Act authorizes the Secretary of Agriculture to carry out operations and measures to detect, control, or eradicate livestock pests and diseases,[Footnote 6] and USDA has delegated this responsibility to its Animal and Plant Health Inspection Service's (APHIS) Veterinary Services. USDA cites this broad authority for implementing NAIS as either a voluntary or mandatory program. NAIS is currently being implemented for nine livestock species groups: bison; camelids (llamas and alpacas); cattle (beef and dairy); cervids (deer and elk); equine (horses, mules, donkeys, and burros); goats; poultry; sheep; and swine. Since 2004, USDA has received input on the design and implementation of the program from various stakeholders, including industry groups, individual producers, livestock markets, slaughter facilities (processors), and state animal health officials. USDA says that because NAIS is a state-federal-industry partnership, the agency has used stakeholder input to adjust the program as NAIS has evolved, and it encourages continued stakeholder input. NAIS consists of three components: (1) registering all "premises" that manage or handle livestock, such as farms, feedlots, veterinary clinics, and livestock markets; (2) identifying livestock animals; and (3) tracking animal movements throughout the production process, from their premises of origin to their slaughter or death. Initially, USDA stated that NAIS would start as a voluntary program and later become mandatory, but, in late 2006, the agency decided that NAIS would remain voluntary. The agency also provided a timeline for implementation and set participation benchmarks that called for gradually increasing the percentages of premises registered, animals identified, and animals tracked. By August 2005, all states had the capability of registering premises, and, as of late May 2007, USDA reported that more than 390,000 premises, or 27.5 percent of the national estimate, were registered in NAIS. Of the total $85.0 million funding made available for NAIS from fiscal years 2004 through 2006, USDA has awarded $35.0 million in cooperative agreements to states, territories, and tribes to help identify effective approaches to register premises and to identify and track animals. In fiscal year 2007, Congress appropriated another $33.0 million to develop and implement NAIS, and the President's Budget in fiscal year 2008 requested an additional $33.1 million for the program. Premises registration is currently funded by USDA, states, territories, and tribes and, therefore, is free to the producer. The costs of animal ID and tracking are to be borne by the livestock industry and will vary, depending on the choices made by individual producers. In this context, we determined (1) how effectively USDA is implementing NAIS and, specifically, the key implementation issues identified by livestock industry groups, market operators, state animal health officials, and others; (2) how USDA has distributed cooperative agreement funding to help states and industry prepare for NAIS and evaluated the agreements' results; and (3) what USDA and others estimate are the costs for USDA, states, and the livestock industry to implement and maintain NAIS. To address all three objectives, we reviewed USDA documents, interviewed agency officials responsible for implementing NAIS, and conducted site visits to selected livestock markets and cooperative agreement field trials. We also conducted structured interviews in person or via telephone with animal health officials in seven states. These states were selected on the basis of their geographic dispersion; the range in the number of premises located in each state; and, in some cases, their high levels of livestock production. We also conducted structured interviews in person or via telephone with, and reviewed documents from, representatives from numerous stakeholder organizations, including several NAIS industry working groups. In addition, for the first and third objectives, we convened a Web-based panel of 32 experts to learn their beliefs and opinions on various aspects of USDA's implementation of NAIS. We selected experts who were actively involved in the development or implementation of NAIS and were knowledgeable of its details; who had conducted research on animal ID, or had published in peer-reviewed journals on animal ID; or who were recognized by their peers as an expert on NAIS. For the second objective, we also reviewed USDA documentation related to cooperative agreements signed between USDA and states, territories, tribes, and industry groups from fiscal years 2004 through 2007. For the third objective, we asked USDA and others for any NAIS cost estimates they had developed, and we reviewed federal guidance for developing cost- benefit analyses. A more detailed description of our scope and methodology is presented in appendix I. We conducted our work from June 2006 to May 2007 in accordance with generally accepted government auditing standards. Results in Brief: USDA has steadily increased the number of livestock premises registered in the nation and taken some steps to address stakeholder concerns in implementing NAIS. However, the agency has not effectively addressed several key issues identified by livestock industry groups, market operators, state animal health officials, and others that, if left unresolved, could undermine the program's goal of rapid and effective traceback and thus hinder its success. Foremost among these issues is USDA's decision in late 2006 to continue implementing NAIS as a voluntary program and to drop participation benchmarks that were intended to gauge progress. Many industry groups, state animal health officials, and experts say this approach may affect the agency's ability to attract the necessary levels of participation to quickly and efficiently locate all animals that are potentially exposed to a disease. However, some industry groups oppose the program being mandatory because they believe that NAIS could succeed as a voluntary program or that USDA first needs to resolve several implementation issues. USDA officials told us that the agency is analyzing what participation levels are necessary to meet the program's goal, and that it may introduce new, risk-based benchmarks, accordingly. In addition, several other key problems hinder the agency's ability to implement NAIS effectively, as follows: * USDA has not prioritized the implementation of NAIS by species or other criteria. Instead, the agency is implementing NAIS for numerous species simultaneously, regardless of the species' economic value, their risk of diseases of concern, the potential human health impact of these diseases, or other criteria. Consequently, federal, state, and industry resources for NAIS have been allocated widely, rather than being focused first on the species of greatest concern and allowing other species to be included later, on the basis of lessons learned. Twenty-one of the 32 expert panel members said USDA should definitely or probably implement NAIS incrementally by species and suggested criteria to prioritize the order of implementation. USDA officials told us that prioritizing implementation may be appropriate, such as focusing on specific diseases of concern or commercial operations, and that the states should determine their own priorities for implementation. * Although USDA aims to minimize the financial and practical impact on producers and others in implementing NAIS, the agency has not developed a plan to integrate NAIS with preexisting animal ID requirements, such as scrapie ear tags and brands, for other USDA and state animal health programs. As a result, producers have generally been discouraged from investing in new ID devices for NAIS, according to industry groups we interviewed. * USDA has not established a robust process for selecting, standardizing, and testing ID and tracking technologies. While international programs have generally used specific animal ID devices for their national animal ID programs, USDA has taken a "technology- neutral" position to allow market forces to determine what devices are most effective and practical. In addition, industry groups, experts, and others told us that electronic ID technologies do not always perform well in production environments, such as livestock markets, and that the agency has not independently tested any ID or tracking devices. Consequently, producers, livestock markets, and others are reluctant to invest in new ID or tracking devices for NAIS, according to industry groups and the experts. * USDA does not clearly define the time frame for rapid animal disease traceback. The definition of "rapid traceback" may vary by disease because some diseases spread more quickly than others, but by not clearly defining a rapid response for a given disease, there could be a slower response and greater economic losses. A senior USDA official told us the agency first needs to identify current baselines for traceback before the agency can determine time-sensitive traceback goals for NAIS. * USDA does not require potentially critical information--such as the species, date of birth, or approximate age of animals--to be recorded in NAIS animal ID and tracking databases. This information can be critical for efficient traceback because it helps limit the scope of an investigation, thus saving time and potentially minimizing the economic impact. USDA officials told us that although animal-specific data can be valuable, the agency is collecting the minimum amount of information needed due to some producers' concerns about protection of their proprietary information in NAIS databases. USDA awarded 169 NAIS cooperative agreements totaling $35 million to 49 states, 29 tribes, and 2 territories from fiscal years 2004 through 2006 to help identify effective approaches to register premises and identify and track animals. To date, USDA has not consistently monitored cooperative agreements, and, as a result, the agency cannot be assured that the agreements' intended outcomes have been achieved. In addition, USDA has not formally evaluated or consistently shared the results of cooperative agreements with state departments of agriculture, industry groups, and other NAIS stakeholders, which would enable lessons learned and best practices to inform the program's progress. USDA officials told us the quality of reports submitted to the agency varies, and USDA has had insufficient resources to conduct additional oversight. In fiscal year 2007, USDA plans to increase oversight of all cooperative agreements awarded that year by assessing progress midyear. Furthermore, USDA plans to give those states with greater numbers of premises registered some flexibility in using cooperative agreement funds to subsidize the purchase of animal tracking equipment for livestock markets. For the first time, in fiscal year 2007, USDA also plans to award $6 million in cooperative agreements to nonprofit industry and other groups to increase premises registration efforts. Although USDA began to implement NAIS in 2004, no comprehensive cost estimate or cost-benefit analysis for the implementation and maintenance of NAIS currently exists. As a result, it is not known how much is required in federal, state, and industry resources to achieve rapid and effective traceback or whether the potential benefits of the program outweigh the costs. Twenty-nine of the 32 expert panel members said that USDA should definitely or probably publish a cost-benefit analysis for NAIS. The NAIS working groups, other livestock industry representatives, and state animal health officials we interviewed also said that the cost of implementing NAIS remained one of their biggest concerns. USDA officials plan to hire a contractor to conduct a cost- benefit analysis, in part, to more precisely forecast the program's economic effects. Moreover, the Senate Committee on Appropriations and the House of Representatives have raised concerns over how USDA has spent funds to develop and implement NAIS. Finally, the experts had mixed views on the impact that NAIS would have on the livestock industry, such as whether NAIS may lead to changes in market structure or affect prices. To ensure that USDA continues to take steps to address unresolved issues, we are making several recommendations aimed at improving USDA's efforts to implement NAIS more effectively and efficiently. For example, we are recommending that USDA reestablish participation benchmarks to gauge progress in registering premises and identifying and tracking animals; monitor participation; and, if participation does not meet the benchmarks, take further action, such as making participation mandatory or creating incentives to achieve those levels of participation. In addition, we are recommending that USDA establish a robust process to select, standardize, and independently test and evaluate the performance of animal ID and tracking devices to ensure they meet minimum standards. We are also recommending that USDA increase the monitoring of NAIS cooperative agreements, evaluate and publish the results of cooperative agreements on a timely basis, and publish the planned analysis of the costs and benefits of NAIS following criteria established in Office of Management and Budget (OMB) guidance. In commenting on a draft of this report, USDA stated that it appreciated our comprehensive evaluation of NAIS and generally agreed with our recommendations. However, regarding our recommendation that USDA establish a robust process to select, standardize, and independently test and evaluate the performance of animal ID and tracking devices to ensure they meet minimum standards, USDA believed that these standards must be defined through a consensus of affected stakeholders and that working with stakeholders to resolve this issue is imperative before selecting specific technologies for NAIS. We recognize the need for USDA to work with stakeholders before determining which ID and tracking devices are most appropriate for NAIS. However, we emphasize that the sooner USDA selects specific technologies, the sooner the animal ID and tracking components of the program will be implemented effectively and efficiently. See the "Agency Comments and Our Evaluation" section and appendix VII for a reprint of USDA's comment letter and our responses to these comments. Background: The concept of animal ID is not new, in the United States or abroad. For decades, American producers have kept records on, and used ID methods for, livestock animals for both commercial and regulatory purposes. Specifically, several USDA and state animal disease eradication programs--such as programs for tuberculosis (TB) in cattle, pseudorabies in swine, and scrapie in sheep and goats--include animal ID requirements. Certain species and classes of animals require officially recognized ID devices for interstate commerce, and all live animals imported into, or exported from, the United States require official ID. Thus, many livestock animals are already identified in the United States by ear tags, branding, tattoos, or other devices. However, the use of ID devices varies by breed, species, and state, and, until NAIS, no attempt had been made to create a uniform animal ID system of national scope and across multiple species using a universal numbering system and central data repository. Due to serious concerns about the United States' ability to safeguard its livestock from the harmful effects of disease, in 2002, the National Institute of Animal Agriculture--an organization of producers, veterinarians, scientists, government representatives, and allied industries--initiated a state-USDA-industry task force of approximately 70 representatives to create a national animal ID system. In 2003, USDA expanded upon this work and established a development team consisting of more than 70 industry associations, organizations, and government agencies. That team ultimately produced the United States Animal Identification Plan in December 2003, which provided the foundation for NAIS. Although early versions of the plan focused on food animals only, other livestock species were later incorporated. The plan was being finalized when the nation's first case of BSE was confirmed on December 25, 2003. Five days later, the Secretary of Agriculture announced measures to guard against BSE and indicated that USDA would expedite the implementation of a national animal ID system. Since 2004, USDA has solicited public comments on draft NAIS policy documents, held public listening sessions, and met with industry groups and others in its efforts to design and implement NAIS. In addition, USDA has received input from 10 working groups comprising producers, academics, and others representing the various livestock species and industry sectors currently included in NAIS. These working groups make recommendations to the NAIS Subcommittee, a group of state and industry stakeholders established by USDA in September 2004 to provide regular, formal input to the USDA Secretary's Advisory Committee on Foreign Animal and Poultry Diseases (full committee) about how NAIS should progress. The NAIS Subcommittee is also responsible for receiving input from the National Institute of Animal Agriculture; the United States Animal Health Association, an organization representing state veterinarians and allied industry groups; and other organizations and individuals. The full committee is a federal advisory group of state, academic, and industry experts selected by the Secretary, which meets once a year or as deemed necessary by the Secretary. While the NAIS Subcommittee meets periodically, the full committee has met only twice- -in September 2004 and September 2006--since the creation of NAIS. USDA is not obligated to take action on the NAIS Subcommittee's, full committee's, or others' recommendations. For the premises registration component of NAIS, states and tribes are responsible for submitting premises information--given to them by producers and others--to a central, Web-based USDA premises database. In turn, USDA allocates a unique, 7-digit, alphanumeric, premises ID number (e.g., A123B45). The premises then receives confirmation online or by mail from the relevant state or tribe with its unique premises ID number. For the animal ID component, USDA published an interim rule, effective in November 2004, recognizing the Animal Identification Number as a new, official numbering system for individual animals in interstate commerce.[Footnote 7] Producers and other NAIS participants can order ID devices from USDA-approved managers that are imprinted with a unique, 15-digit Animal Identification Number for use on animals that move through the production process as individuals, as is typical in the cattle, sheep, and goat industries. Figure 1 shows a calf that is identified with tags in both ears, with the calf's left ear bearing an electronic tag and its right ear bearing a visual tag. Both tags have the Animal Identification Number for official ID purposes, and the visual tag also has a number used for the producer's herd management purposes. For animals of the same species that typically move through the production chain as a group, such as commercial poultry and swine, producers can instead identify the animals with a group/lot ID number. Group/Lot ID numbers are self-generated by the premises (not assigned by USDA) and are maintained at the premises in management records. Figure 1: A Calf Identified with Both a Visual and an Electronic Tag: [See PDF for image] Source: USDA. [End of figure] Lastly, for the animal tracking component, USDA has developed the Animal Trace Processing System to allow state and federal animal health officials to request information, in the event of an animal health investigation, from multiple private and state animal tracking databases containing animal location and movement records. Producers and others are responsible for reporting certain animal movements, such as when a change of ownership occurs or when animals commingle with other herds or flocks. Events that would enable state and federal animal health officials to request information from this network of databases include an indication or a confirmed positive test of a foreign animal disease, an animal disease emergency as determined by the Secretary of Agriculture or state departments of agriculture, or a need to conduct a trace to determine the origin of infection for a domestic disease of concern. (See app. II for a list of domestic and foreign animal diseases of concern identified by USDA.) Key NAIS Implementation Issues Are Unresolved and Could Undermine the Program's Goal of Rapid and Effective Traceback: USDA has steadily increased the number of livestock premises registered in the nation and has taken some steps to address stakeholder concerns in implementing NAIS. However, the agency has not effectively addressed a number of key issues identified by livestock industry groups, market operators, state animal health officials, and others that, if not ultimately resolved, could undermine the program's goal of rapid and effective traceback, thus hindering its success. USDA's decision to implement NAIS as a voluntary program without benchmarks to measure progress may affect the agency's ability to attract the necessary levels of participation to quickly and efficiently locate all animals potentially exposed to a disease. In addition, USDA has not prioritized the implementation of NAIS by species or other criteria. Furthermore, USDA has not developed a plan for integrating NAIS with other USDA and state animal ID requirements, nor has it established a robust process for selecting, standardizing, and testing ID and tracking technologies. The agency also does not clearly define the time frame for rapid traceback. Finally, USDA does not require potentially critical information for efficient traceback to be recorded in NAIS databases. USDA Is Implementing NAIS as a Voluntary Program without Participation Benchmarks: During the first 2 years of the program's implementation, USDA stated several times that participation in NAIS would initially be voluntary but would eventually become mandatory to achieve full participation and, thus, the goal of rapid and effective traceback. One of USDA's first major NAIS policy documents, the Draft Strategic Plan 2005-2009, released in April 2005, stated that during initial implementation, participation would be voluntary so that stakeholders could have the opportunity to obtain experience with the program and provide feedback as successful and practical solutions evolved. The plan also provided a timeline for implementation, with premises registration and animal ID to be required by January 2008 and the reporting of defined animal movements to be required by January 2009, under what would become an entirely mandatory program. The plan stated that this phased-in approach was "to support the transition from voluntary to mandatory as full implementation is achieved." In addition, the plan also stated that, "While market forces may eventually create more inclusiveness, the clear stakeholder support for transitioning to a mandatory program and the urgency of achieving the goal, suggest that setting a date for that transition would benefit the program." Later, the April 2006 Strategies for the Implementation of NAIS set benchmarks that were intended to gauge progress in attaining full participation. These benchmarks called for gradually increasing the percentages of premises registered, animals identified, and animals tracked from January 2007 through January 2009. Specifically, for premises registration, USDA aimed for 25 percent participation by January 2007, 70 percent by January 2008, and 100 percent by January 2009. In addition, the implementation plan called for 40 percent of animals being identified by January 2008, 100 percent of "new" animals less than 1 year of age being identified by January 2009, and 60 percent of new animals having complete tracking data by January 2009. USDA stated that it would evaluate whether participation levels were increasing at rates that would achieve full participation by 2009, and that, if this were not the case, USDA would develop federal regulations to require industry to identify their premises and animals. In May 2006, in an announcement for NAIS cooperative agreements, USDA stated that it anticipated promulgating regulations by early 2008 to require participation in all three components of the program. Furthermore, in a document supporting its fiscal year 2007 budget submission to OMB, USDA stated that the intrinsic value of its investment in NAIS was based on the assumption that there would be sufficient animal ID and movement data available to support the program's traceback goal. USDA also said that if participation fell below expectations, traceback would periodically fail because of the lack of animal tracking information from nonparticipants. However, in August 2006, reporting that 20 percent of the nation's premises had been registered, USDA decided that NAIS would permanently remain a voluntary program. USDA officials told us that due to opposition from some industry groups, it is imperative that industry advance the program, rather than government regulations, to encourage participation. Some industry groups believe that NAIS could succeed as a voluntary program or that USDA first needs to resolve several key implementation issues before making participation mandatory. For example, officials from the National Cattlemen's Beef Association told us they believe that NAIS could be successful as a voluntary program, such as the beef industry's 20-year-old Beef Quality Assurance program, which covers about 95 percent of cattle in feedlots and aims to reduce drug residues and pathogen contamination. The association officials also told us that if NAIS became mandatory, producers who have voluntarily participated would lose the market advantage they currently enjoy through higher prices paid at market or slaughter for animals they identify for marketing or management purposes. Another industry group, the American Farm Bureau Federation, which in 2006 supported NAIS being a mandatory program, expressed its support in January 2007 for NAIS to be voluntary, while cautioning that USDA should not make the program mandatory until the agency has published a full cost analysis for the program. The Livestock Marketing Association--a national trade association representing over 700 livestock auction markets, dealers, and other livestock marketing businesses--also wrote in comments to USDA in 2005 that NAIS should remain voluntary until USDA addresses several implementation issues, including the effectiveness and availability of animal ID technology. In November 2006, USDA also dropped its participation benchmarks from the most recent policy document for the program, the draft NAIS User Guide.[Footnote 8] Despite a steady increase in the number of premises registered since USDA began reporting premises registration information in January 2005 and reaching the 25 percent target (nearly 360,000 premises) in early February 2007 (see fig. 2), USDA officials told us they dropped the participation benchmarks because meeting future benchmarks for all components of NAIS was no longer realistic. Instead, the officials said the agency is analyzing what participation levels for all components of NAIS would achieve the "critical mass" necessary to have an efficient and effective program. The officials added that the agency does not expect that equal levels of involvement across all species will be necessary, and that new, risk-based participation benchmarks for premises registration, animal ID, and animal tracking may be developed accordingly, which could vary by species. However, USDA has not determined what action it may take if participation levels do not meet those new benchmarks. Figure 2: Number of NAIS Premises Registered, 2005 to Present: [See PDF for image] Source: GAO summary of USDA data. Note: On the basis of 2002 U.S. Census of Agriculture data, USDA's National Agriculture Statistics Service estimates there are approximately 1.4 million distinct livestock premises nationwide (with more than $1,000 in annual income), which may contain one or more species. [End of figure] Although it may be too soon to determine whether USDA's current approach will be successful, many industry groups, state animal health officials, and the experts we surveyed say the program will likely need to become mandatory to achieve the levels of participation that are necessary to rapidly and effectively locate all potentially exposed animals in a disease traceback. In the Draft Strategic Plan, for instance, USDA reported that most individuals who spoke about this subject at the agency's 2004 listening sessions preferred, by a ratio of 3:1, a mandatory program to a purely voluntary program. USDA also reported at that time that a survey of National Institute of Animal Agriculture members showed even stronger support, by a ratio of 8:1, for a program that is or will become mandatory. In addition, state and industry officials we interviewed said that as a voluntary program without benchmarks, NAIS has lost momentum, deterred participation, and faces an uncertain future. For example, officials from one major agricultural state told us that USDA's changed direction on whether the program would become mandatory has challenged the state's premises registration efforts, because many producers are motivated by compliance with federal requirements, not necessarily by NAIS's traceback goal alone. These officials also said that the lack of participation benchmarks had compromised the state's credibility with producers and its ability to make progress in implementation. As of early June 2007, this state had less than 16 percent of its premises registered, and the officials believed there is little incentive for producers to participate in NAIS. Moreover, three states where premises registration is mandatory by state law--Indiana, Michigan,[Footnote 9] and Wisconsin--accounted for about 26 percent of USDA's total premises registered nationally as of that time. Furthermore, a majority of the 32 expert panel members said that 81 percent to 100 percent of producers, livestock markets, and slaughter facilities would need to register their premises to achieve the program's goal of rapid and effective traceback (see apps. III and IV for more details about these experts and their responses to our questions). By contrast, under a voluntary program, only 1 expert believed that producers would achieve at least 81 percent participation in premises registration, while 11 experts believed that level would be achieved by livestock markets, and 20 experts believed that level would be achieved by slaughter facilities. For the animal ID and tracking components, experts had similar views. For example, a majority believed that 76 percent to 100 percent of producers, markets, and slaughter facilities would need to participate in animal tracking to make the program effective, while a minority believed that level would be achieved in any of those sections under a voluntary program. Overall, 27 of the 32 experts said participation in NAIS should definitely or probably be mandatory. Several other countries, including the United States' major agricultural trading partners and competitors, have instituted mandatory animal ID programs for cattle and, in some cases, a few other species. For example, the EU has mandatory programs in which all cattle born or moved across EU state lines as of 2000 must be identified with two individual ear tags and an animal passport, and member states must maintain computerized databases that record births, movements, and deaths. Since several cases of BSE were discovered in 2001, Japan has made a series of changes to its food safety legislation, resulting in a mandatory system where all beef and dairy cattle must be identified using an ear tag. Information is maintained on an animal's ID number, breed, gender, and production history from the farm of origin through distribution to consumers. Similarly, in 2001, Canada started a compulsory animal ID program that applies to all bovine and bison and now requires that animals receive a radio frequency identification (RFID) tag when leaving their herd of origin,[Footnote 10] which is collected at slaughter or export. Canada later expanded its program to sheep in 2004, requiring the use of visual ear tags. Brazil, the world's largest beef exporter, gradually phased in its mandatory ID program for cattle and bison starting in January 2002; the entire herd is expected to be identified by the end of 2007. Australia, the world's second-largest beef exporter, has developed a mandatory system that uses RFID to identify and trace cattle from farm of origin to slaughter. Australia has been moving toward a fully integrated program linking electronic ID devices, product bar coding, and a central electronic database. Appendix V provides more detailed information on select international animal ID and tracking programs.[Footnote 11] To increase participation in NAIS, several industry groups, state animal health officials, and the experts we surveyed have also suggested that USDA provide incentives, such as financial assistance, to industry to implement the animal ID and tracking components. For instance, the NAIS Cattle Working Group recommended in 2004 that USDA assume primary financial responsibility for funding the initial development of the basic infrastructure required for animal ID and tracking--including equipping concentration points, such as livestock markets and slaughter facilities, with RFID readers and software to capture the data electronically. Livestock market operators and others also say such financial support would be an attractive incentive because RFID technology, if effective, would allow animal movements to be recorded as quickly as the current "speed of commerce" and, therefore, would not slow down business operations. The NAIS Subcommittee also recommended, in 2005, a cost-sharing arrangement between USDA and industry to fund the program's implementation. The Secretary's Advisory Committee on Foreign Animal and Poultry Diseases adopted these recommendations as well as the other NAIS working group and Subcommittee recommendations presented at its September 2006 meeting. One state that has already created a cost-sharing incentive is Wisconsin, where the state contributes 50 percent, or up to $1, of the cost of an RFID ear tag on a first-come, first-served basis under its voluntary animal ID program. Wisconsin officials say the cost-sharing arrangement is designed to make the program more attractive to the state's producers, and, as of early March 2007, the state had approved reimbursing producers for a total of 30,000 tags for cattle, which was up from 20,000 tags the previous month. However, USDA officials told us that the power of producers to protect themselves and their animals is a tremendous incentive for participation in NAIS, and, as the program continues to be implemented and developed, additional incentives will be realized by USDA, producers, and state and industry partners. Nonetheless, whether NAIS is mandatory or voluntary, the lack of participation benchmarks prevents USDA from measuring progress in attaining the necessary participation levels for an effective program. Several Other Key Issues Hinder USDA's Ability to Implement NAIS Effectively: Industry groups, market operators, state animal health officials, and others have identified several other key problems that, if left unresolved, could undermine the program's goal, further hindering USDA's ability to implement the NAIS program effectively. USDA Has Not Prioritized the Implementation of NAIS by Species or Other Criteria: USDA has not prioritized the implementation of NAIS by species or other criteria. Instead, the agency is currently implementing NAIS simultaneously for numerous species, regardless of their economic value, their risk of diseases of concern, the potential human health impact of these diseases, or other criteria. Consequently, federal, state, and industry resources for NAIS have been allocated widely, rather than being focused first on the species of greatest concern and allowing other species to be included later, on the basis of lessons learned. In contrast, international animal ID programs have generally started implementation with one species (cattle) and, in some cases, later expanded to include a few other species, such as sheep, bison, and goats, on the basis of disease risk, economic importance, or potential human health impact. In addition, 21 of the 32 expert panel members said USDA should definitely or probably implement NAIS incrementally by species, while 8 experts said USDA should definitely or probably continue with its current approach to implement the program for numerous species simultaneously.[Footnote 12] Many of the 32 experts suggested criteria that USDA could use to determine the priorities given to each species in implementing NAIS, including whether it is a food animal; its likelihood of capturing diseases of concern; the risk that the animal will spread a disease harmful to human health; the relative ease of implementation for a particular industry; and the industry's impact on the U.S. economy, including export value. Prioritizing implementation for certain species before trying to apply the program to numerous species would likely result in more efficient and cost-effective implementation, on the basis of lessons learned and best practices being identified. In addition, the industries currently covered by NAIS appear to be at various stages of readiness to implement NAIS. For example, the Bison, Camelid, Cattle, Equine, Swine, Sheep, and Goat Working Groups have submitted reports to USDA with implementation recommendations for their industries, but the Cervid and Poultry Working Groups have not. Furthermore, according to industry representatives, some industries, such as commercial swine and poultry operations, already have widespread ID and tracking systems in place that would allow traceback to occur within 48 hours, while other industries may not. USDA officials pointed out that an animal ID program to support the animal health needs of all livestock species would be unique in the world and would place the United States in a position to set a new standard for animal ID. Some industry groups have expressed their support for NAIS being implemented for all species, since many transmissible diseases are not species-specific. For example, the National Livestock Producers Association, which represents about 200,000 producers across the country, wrote USDA in 2005 that the true value of NAIS rests in its potential ability to track all livestock animals, regardless of species, due to the extent of their commingling and potential to spread disease. The association added that NAIS would not be very effective or equitable if all species were not included as soon as possible. Furthermore, 21 of the 32 experts we surveyed said USDA should definitely or probably continue with its current approach to include all species, rather than limit NAIS to one or a few species. USDA officials also told us that establishing NAIS across all species is critical, because many operations handle more than one species, and that focusing entirely on cattle, as some other countries have done, is a critical flaw since some animal diseases cross species lines. Nonetheless, USDA officials told us they recognize that prioritizing program implementation may be appropriate, such as by focusing on specific diseases of concern or large commercial operations, and that the states should determine their own priorities for implementation. These officials also said future NAIS plans will more clearly identify higher-risk areas or sectors within the species. In commenting on a draft of this report, USDA told us it plans to develop, in collaboration with the species working groups, a NAIS Short-Term and Long-Term Implementation Strategies document that will contain actions for the remainder of 2007 through 2011. Specifically, the agency stated that the short-term strategy, targeted for publication in August 2007, will target species or industry sectors that have the greatest need for advancing premises registration, animal ID, and tracking. The long-term strategy will be distributed in early 2008 and will call for an evaluation of participation through 2009 to determine what actions, such as incentives, may be needed to accelerate participation in the voluntary program. USDA Has Not Developed a Plan for Integrating NAIS with Other USDA and State Animal ID Requirements: Although USDA aims to minimize the financial and practical impact on producers and others in implementing NAIS, the agency has not developed a plan to integrate NAIS with preexisting programs and systems. Many producers are already required to participate in preexisting USDA and state animal disease eradication programs that use specific ID devices with different numbering systems or that require branding. For example, the National Scrapie Eradication Program for sheep and goats requires visual ear tags, and other USDA-state programs require ear tags for brucellosis and TB in cattle and ear notches in swine. In addition, several western states recognize branding as an official ID for disease control purposes. As a result, producers have generally been discouraged from investing in new ID devices for NAIS, according to industry groups we interviewed, thereby inhibiting implementation of the program's animal ID and tracking components. Importantly, while USDA published an interim rule effective in November 2004 recognizing the Animal Identification Number as a new, official numbering system for individual animals in interstate commerce, this new system does not replace other, USDA-recognized, official numbering systems. The rule established that the Animal Identification Number may be used for official ID in other disease eradication programs. Nonetheless, USDA officials told us that they are evaluating how NAIS and other official ID systems can be standardized and moved to a single numbering system, to the extent practical, and that eventually, the agency expects Animal Identification Numbers to become the standard national numbering system used for certain species, individual ID methods, or both. USDA told us that as of early March 2007, 1.3 million Animal Identification Number RFID ear tags had been distributed, including some that are being used for state disease eradication programs. For example, USDA reported that about 500,000 tags had been distributed to Michigan producers, where the state's bovine TB eradication program requires all cattle to have RFID ear tags prior to movement from their premises. For NAIS not to impose undue costs on producers by requiring additional ID devices, stakeholders say the program must be integrated with preexisting programs and systems. However, USDA faces challenges in integrating NAIS with other animal ID requirements. For example, Sheep Working Group members told us that because the scrapie program already assigns a flock ID number to each premises--plus a unique, individual ID number to each animal--sheep producers do not see the need to participate in NAIS, which involves different premises and individual animal ID numbering systems. Another challenge is that brands identify all animals raised by a specific producer as a group, not as individuals, and the same brands are often used in different states or even in different counties within the same state. USDA's NAIS User Guide states that registered brands are not considered to be an official, individual animal ID as called for by NAIS because cattle typically move through the production process as individuals. In addition, NAIS animal tracking requirements may differ operationally from state brand laws and practices. For example, New Mexico requires state authorities to inspect all livestock moving across brand district lines, which contrasts with NAIS, where the responsibility of reporting animal movement lies with the premises receiving animals. New Mexico officials told us that if they were to implement NAIS as envisioned, the time required for state inspections would at least double if inspectors were required to read and report ID tags. The Cattle Working Group recommended, in 2004, that USDA develop protocols for integrating existing brand laws with NAIS individual animal ID requirements and for the reporting of animals' movements from brand law states to nonbrand law states. USDA officials told us that the need to have a single numbering system across all species is less important than getting animals individually identified and tracked using any official ID system recognized by USDA. Furthermore, these officials said that while standardizing to a single numbering system for animal ID may eventually be appropriate, USDA recognizes there are differences among species and that cost, technology capability, and practicality must be considered before phasing out existing ID devices that have proven to be workable for producers. For example, the 15-digit Animal Identification Number may not be the most practical numbering system when used on visual ear tags for smaller animals since the size of the tag does not lend itself to a 15-digit number. Nonetheless, USDA officials are starting to address some integration issues. For example, USDA officials told us in April 2007 that the agency has decided to allow the official ID devices and numbering systems used by other disease eradication programs for the purposes of NAIS as well, although the agency has not yet communicated this development to industry. In addition, in late 2006, a NAIS Brand State Working Group was formed, in part to identify what brand concepts could integrate with NAIS. In commenting on a draft of this report, USDA informed us that it will update the NAIS User Guide in October 2007 to more clearly reflect the use of other official ID numbers within NAIS. USDA Has Not Established a Robust Process for Selecting, Standardizing, and Testing ID and Tracking Technologies: International programs have generally used specific animal ID devices for their national animal ID programs, and some NAIS working groups have recommended specific ID devices for their species, such as RFID ear tags for cattle and RFID microchip implants for horses. However, USDA has taken a "technology-neutral" position to allow market forces to determine what devices are most effective and practical and to accommodate future technologies. In a NAIS policy document on ID devices released in February 2006,[Footnote 13] USDA stated that individual, visual ID devices are a starting point to ensure greater participation among producers and asserted that a neutral approach allows RFID; biometrics, such as DNA and retinal imaging devices; and other potential technologies to be used as supplemental identification. Nonetheless, that document also stated that uniformity and compatibility of technology are critical to ensure that the collection of animal ID data is practical and cost-effective throughout production. Furthermore, USDA recognized the need to have ID technologies that are compatible with Canada and Mexico. USDA also stated that as NAIS is phased in, ongoing efforts to harmonize animal ID with other countries will facilitate safe trade. While not all species can use the same devices due to industry preferences or physical limitations, such as small ears, USDA's technology-neutral approach means, for example in the cattle industry, that a producer can choose to use visual ear tags; low-or high- frequency RFID ear tags; or other advanced technologies, such as retinal imaging. There are costs and benefits associated with any device. For example, visual ID devices are less expensive but require manual recording, which may cause errors and slow down the "speed of commerce" at livestock markets and slaughter facilities. RFID systems, on the other hand, allow data to be captured automatically into databases, but these systems are also not consistently accurate and are more expensive--in terms of both the ID device and the associated infrastructure (reader, installation, and computer use). With such a wide range of options in animal ID and tracking devices, industry groups and expert panel members told us that producers and market operators fear that their choices may be inconsistent with others in the marketplace, or that USDA will adopt specific devices in the future, and they may find themselves having made the wrong investment decision. For instance, a producer may find that the closest livestock market uses electronic readers and cannot easily accommodate visual ear tags; alternatively, the market may not have installed RFID reader equipment, and the producer would not get the anticipated return on his or her investment. From another perspective, a multispecies livestock market, based on its customers, may face a dilemma of investing in equipment to read and record visual tags, RFID tags, RFID implants, and other devices--or risk being unable to capture all information quickly and efficiently and losing some customers. Consequently, producers, livestock markets, and slaughter facilities have generally been discouraged from investing in ID or tracking devices, thus inhibiting implementation of the animal ID and tracking phases. Asked whether USDA's technology-neutral position encourages or discourages producers' investment in animal ID technology, 23 of the 32 expert panel members said this position definitely or probably discourages investment, and 6 said it definitely or probably encourages investment.[Footnote 14] In their written responses elaborating on this question, several experts said USDA's technology-neutral approach limits the interoperability (compatibility) of different systems in place, thus reducing the viability of a consistent, national traceback program. In addition, they wrote that it has caused confusion, uncertainty, and a "wait-and-see" attitude in the marketplace, and that it will take time to sort out efficient from inefficient technologies. Conversely, other experts replied that USDA's technology-neutral approach allows marketplace competition to advance new or improved technologies and drive fair prices. Similarly, for animal tracking, more experts replied that USDA's approach definitely or probably discourages investment by producers, livestock markets, and slaughter facilities more than it encourages investment. For example, 22 experts said USDA's approach discourages investment by livestock markets, compared with 2 who said it encourages such investment; 17 experts said it discourages investment by slaughter facilities, whereas 4 said it encourages that industry sector to invest. While USDA has not selected specific animal ID devices, the agency has published minimum standards for the various ID devices recommended to date by the species working groups and the NAIS Subcommittee. Specifically, USDA has established printing and performance standards for visual and RFID ear tags that address characteristics such as durability (expected tag life); tag loss; visual readability of the 15- digit Animal Identification Number; and, in the case of RFID, electronic read rates and ranges. USDA has published similar performance standards for RFID implants, with additional characteristics addressing the migration or breakage of the device and its being harmless to an animal. However, USDA has not published standards for RFID readers and does not expect to do so unless the agency purchases readers for use by animal health officials, in which case it says it will define performance standards for those specific environments. In addition, USDA has not established a robust process to independently test and evaluate the performance of animal ID and tracking devices. Industry groups, expert panel members, and others told us that RFID devices do not always perform well in production environments, such as livestock markets, particularly with RFID readers being made by different companies. For example, one NAIS pilot project found that in loading cattle onto commercial trucks, RFID readers read only 70 percent of the RFID ear tags, with variations among tag manufacturers ranging from 47 percent to 96 percent.[Footnote 15] As a result, stakeholders are lacking reliable, independent information on the effectiveness of animal ID and tracking devices, and without such information, they are reluctant to invest in these devices. It is common U.S. practice to select one technology for systems that need to be widely implemented in different environments to ensure consistency and interoperability across multiple users. We have previously reported that a robust process for selecting technologies, setting and revising performance standards, and testing and evaluating technologies against those standards leads to the most effective and efficient use of technology.[Footnote 16] For example, we have reported on the necessity of the federal government's selection and standardization of RFID cards and readers for federal employees so that the ID cards can be read at any federal agency across the nation.[Footnote 17] Several expert panel members suggested that USDA provide funding for independent, third-party evaluations of technologies and make results of such work readily available; appoint a standards committee to evaluate and make recommendations on the basis of sound science; or require independent evidence that devices meet standards before approving them. The NAIS Subcommittee also recommended, in 2006, that USDA establish an objective process to (1) test the performance of ID devices to ensure they meet NAIS standards in various production environments and over extended periods and (2) evaluate new technologies as they emerge. Typically, federal agencies rely on independent laboratories that are certified by a government agency, such as the National Institute of Standards and Technology, for such testing and evaluation. USDA officials told us they are working with industry to determine better ways to define performance criteria and establish a more thorough process to test and evaluate ID devices, but they did not specify a time frame for these developments. Toward this end, the agency held preliminary discussions in mid-April 2007 with the American Society for Testing and Materials' Committee on Livestock, Meat and Poultry Evaluation Systems to form a task force to fine-tune ID performance standards for NAIS and help USDA put testing protocols in place. In the February 2006 NAIS policy document on ID devices, USDA stated that when NAIS becomes fully operational, the agency will develop an approval process for official ID devices and more complete testing and evaluation procedures. Manufacturers of Animal Identification Number devices, regardless of any prior permission from USDA, will have to submit new or appended applications to be considered for "USDA Approved" status. Evaluations may include laboratory or field studies to verify compliance with criteria and specification standards, either before or following issuance of "USDA Approval Pending" or "USDA Approved" status for ID devices. USDA Does Not Clearly Define the Time Frame for Rapid Traceback: When USDA announced NAIS in 2004, the program's traceability goal was to locate all potentially exposed animals within 48 hours of a disease's discovery, and both USDA and states conveyed that message in their outreach to producers and others. However, USDA's most recent NAIS policy document, the draft NAIS User Guide issued in November 2006, is silent on this time frame and instead says NAIS will allow producers and animal health officials to respond as "quickly, efficiently, and effectively as possible." By definition, traceback goals need to be time-sensitive and cost-effective to efficiently target and evaluate the program's success in eliminating a disease outbreak. If rapid traceback goals are not clearly defined, there could be a slower response to an animal disease outbreak and, therefore, greater economic losses. Of the 32 expert panel members, 25 defined rapid traceback in an animal disease event as occurring within 48 hours, with 10 of the experts defining it as 24 hours or less and 15 defining it as 25 to 48 hours. State animal health and industry officials told us it is important that USDA communicate a specific time frame to encourage participation, reinforce the necessity of rapid traceback, and have a measure by which to evaluate results. However, a senior USDA official told us that the definition of "rapid traceback" may vary by disease, because some diseases spread more quickly than others and some diseases are limited in how they can be transmitted. For example, traceback for FMD might ideally occur within 12 hours because the disease spreads so rapidly. By contrast, because BSE is transmitted only through animal feed containing certain contaminated animal products, and scrapie is transmitted during the breeding season, a longer traceback would be appropriate. In addition, the senior official told us that until USDA collects baseline information on tracebacks for specific diseases, the agency cannot determine time-sensitive, cost-effective traceback goals for NAIS. USDA Does Not Require Potentially Critical Information for Efficient Traceback to Be Recorded in NAIS Databases: When producers and other participants register their premises, they are required by USDA to record only their name and contact information, with species information being optional. In addition, when USDA- approved managers distribute ID devices to a producer, they must record the devices' unique animal ID numbers, the premises ID number where the devices were sent, and the date of distribution. However, USDA does not require additional information, such as the species, date of birth, or approximate age of the animals, to be recorded in NAIS animal ID or tracking databases. Information that may be critical for narrowing the scope of a traceback--thus saving time and resources and potentially minimizing the economic impact--includes the species, date of birth, or approximate age of an animal. Many diseases, such as bovine TB, affect only specific species or generally affect animals of a certain age, such as in the case of Johne's disease, which is usually contracted at a young age. Consequently, if a new case of these diseases arose in the United States, tracing other species or animals of a different age may unnecessarily use federal, state, and industry resources in locating animals and premises that may not be affected--thus impeding the goal of rapid and effective traceback. Similarly, most equine diseases of concern affect only equine species, and exotic Newcastle disease affects only poultry, so tracing other species would be an inefficient use of time and resources. Other state and federal animal disease eradication programs require the recording or reporting of this type of information. For example, the Bovine Tuberculosis Eradication Program requires, for all TB-tested bison and cattle, the reporting of the animal's unique, official ID device; approximate age; gender; and breed. In addition, the National Scrapie Eradication Program requires goat flock owners to maintain a management and monitoring plan that must record an animal's gender, year of birth, and breed following the discovery of scrapie within the flock. The Cattle Working Group recommends that producers identify calves at birth or at the earliest date possible to support animal disease issues when the age of an animal is needed, noting that when the precise date of birth is not known, the approximate birth date within 2 to 3 months should be recorded. USDA officials acknowledged that although animal- specific data can be valuable, the agency is collecting the minimum amount of information needed for traceback to (1) respond to some producers' concerns about protection of their proprietary information in NAIS databases and (2) encourage participation. In addition, participants have the option to record such information in ID and tracking databases, and USDA encourages them to do so. Nonetheless, without this information being consistently recorded in NAIS databases, USDA and state officials may not be able to efficiently trace only those animals potentially affected by a disease. USDA Has Awarded NAIS Cooperative Agreements to Identify Effective Implementation Approaches but Has Not Formally Evaluated Agreements' Results: USDA has awarded $35.0 million in NAIS cooperative agreements to states, tribes, and territories to help register premises and identify and track animals. However, USDA has not consistently monitored or formally evaluated the results of these cooperative agreements. In addition, USDA has not consistently shared cooperative agreement results with NAIS stakeholders. USDA plans to increase its oversight and give states with greater participation in NAIS some flexibility in using their cooperative agreement funds. USDA Awarded 169 Cooperative Agreements between Fiscal Years 2004 and 2006: To help identify effective approaches to register premises and identify and track animals between fiscal years 2004 and 2006, USDA awarded 169 cooperative agreements, totaling $35.0 million, to 49 states, 29 tribes, and 2 territories. NAIS cooperative agreement awards ranged in size from $7,381 to $1.2 million, and the average award was about $207,000. In fiscal years 2004 and 2005, USDA did not require recipients to contribute to, or share, costs (cost-share); however, in fiscal years 2006 and 2007, USDA required certain recipients to demonstrate 20 percent in matching funds, through cash or in-kind (noncash) contributions. For more information about NAIS cooperative agreements' funding by fiscal year, see appendix VI, table 2. Cooperative agreements are typically funded for a 12-month funding period, with recipients required to submit both quarterly accomplishment and financial status reports. USDA required all NAIS cooperative agreement recipients to submit information to USDA in an initial work plan containing proposed project objectives, species and industry sector focus, as well as milestones for measuring progress. Although some premises registration cooperative agreements proposed activities that would span across most species and industry sectors covered by NAIS, other projects proposed focusing premises registration activities on one, or a few, species and sectors of the livestock industry. Of field trial cooperative agreement funding to test animal ID and tracking solutions, all but two field trials intended to focus on beef or dairy cattle; several proposed work on sheep; a few included swine, cervids, goats, bison, and equine; and only one project intended to focus some work on camelids. While no NAIS field trials proposed work on the poultry industry, USDA has previously funded work, through funds other than NAIS cooperative agreements, that examined the tagging and record-keeping requirements that would facilitate tracking of birds in the live bird marketing system. For information on species covered under NAIS field trials, see appendix VI, table 3. Field trial funding recipients also varied in the number of industry sectors they intended to involve in cooperative agreement activities. Overall, more than one-half of recipients intended to work with producers, livestock markets, slaughter facilities, and feedlots. USDA did not require field trials covering multiple species or industry sectors to include in initial work plans information regarding how funding was to be spent on each species or sector. USDA officials told us that they have not requested such details because of the interconnectivity of the activities associated with cooperative agreements. Most of the cooperative agreement awards were focused on premises registration. Specifically, of the $35.0 million awarded in fiscal years 2004 through 2006, USDA awarded 146 cooperative agreements totaling $23.4 million for premises registration efforts to 49 states, 29 tribes, and 2 territories. These premises registration awards provided funding for activities such as hiring personnel to register premises, developing educational materials, and providing outreach to producers and nonproducer participants on the goals of NAIS. For example, 1 cooperative agreement awarded to the Navajo Nation was for the development of communications in the Navajo language for outreach on premises registration and animal ID. In addition, some of these cooperative agreements funded limited animal ID and tracking activities, along with premises registration. In fiscal years 2004 and 2005, USDA also awarded field trial cooperative agreements. Specifically, in fiscal year 2004, USDA awarded 16 cooperative agreements totaling $9.7 million to 15 states and 1 tribe. USDA estimates that $1.8 million of the $9.7 million awarded was used to support premises registration activities. The remainder was used for field trials to develop, test, and offer solutions for applying animal ID devices and collecting animal tracking information. For example, 1 cooperative agreement with the Wyoming Livestock Board tested whether existing brand inspection personnel and infrastructure could be used to track livestock changing ownership and livestock entering into interstate commerce through Wyoming livestock markets. In fiscal year 2005, USDA awarded 7 cooperative agreements totaling $1.9 million to 6 states and 1 tribe for field trials to support research, including the assessment of existing and novel ID technologies. USDA anticipates awarding an additional $20.5 million in cooperative agreements in fiscal year 2007. Of this, USDA anticipates awarding $14.5 million for continued support of premises registration, education, and outreach activities under approximately 80 cooperative agreements to 50 states, 28 tribes, and 2 territories. As of March 2007, USDA had awarded 31 of the anticipated 80 cooperative agreements, totaling $6.7 million. The remaining $6.0 million in fiscal year 2007 funding for cooperative agreements will be provided for the first time to nonprofit organizations for premises registration activities. In January 2007, USDA entered into a cooperative agreement with the National Pork Board to begin work with pork producers to encourage premises registration. Subsequently, in February 2007, USDA announced that other nonprofit industry organizations, historically black colleges, tribal land-grant colleges, and tribal organizations were also eligible for these awards, which are intended to support the continued registration of premises. USDA Has Not Consistently Monitored or Formally Evaluated NAIS Cooperative Agreements or Consistently Shared Their Results: To date, USDA has not consistently monitored or formally evaluated NAIS cooperative agreements and has not consistently shared their results with state, industry, and other stakeholders. USDA officials told us that NAIS program staff provided some oversight for field trial cooperative agreements as well as for tribal premises registration cooperative agreements. However, NAIS program staff do not directly monitor most NAIS cooperative agreements; instead, USDA delegates administrative oversight activities for each cooperative agreement to designated representatives, mostly Area Veterinarians in Charge whose overall responsibility is to supervise and perform the official animal health activities of APHIS in the state concerned. These individuals are responsible for direct administration of the individual state premises registration and field trial cooperative agreements, including the monitoring and assessment of agreements. According to USDA, it is appropriate for these designated representatives to monitor cooperative agreements because they are familiar with the circumstances associated with implementing projects in a particular state. We have previously reported with other government audit organizations that monitoring the performance of federal awards helps to ensure that goals are reached and required deliverables are completed.[Footnote 18] According to USDA officials, the designated representatives responsible for monitoring cooperative agreements have a multitude of competing responsibilities and thus may not have sufficient time or resources to oversee cooperative agreements. In addition, NAIS program staff did not conduct any formal evaluation of NAIS cooperative agreements. Evaluating results against cooperative agreement goals can help to identify ways to improve program performance. USDA officials said that the quality of quarterly and final accomplishment reports provided to designated representatives by cooperative agreement recipients varied. They said that many times, these reports identified what was being done, rather than what was being accomplished. For example, one state premises registration project set milestones for educating 60 percent of livestock producers and registering 40 percent of premises; however, the reported results included the purchase of computers, number of presentations given to producer organizations, and number of premises registration forms that were printed for distribution. Furthermore, our analysis of available reports indicated that results were not reported or were of limited value because initial project goals and milestones presented in recipients' work plans were vague or unclear.[Footnote 19] In addition, several states indicated to us that they had difficulty completing the work outlined in their cooperative agreements within given time frames. For instance, one state told us that it was difficult for it to hire an ID coordinator in its first year of cooperative agreement funding, which meant the state was unable to use all of the funds it was allocated. Thus, while USDA has awarded the majority of cooperative agreement funds to support premises registration, the agency has not been able to determine effective or ineffective approaches for increasing premises registration, animal ID, or tracking. USDA has formally shared few results of cooperative agreements with NAIS stakeholders, hindering them from identifying approaches that have worked to achieve NAIS program objectives, such as increasing outreach, as well as ineffective approaches, such as the interoperability of RFID devices and readers. To date, USDA has provided information regarding NAIS field trial cooperative agreements on three occasions. In April 2005, USDA released a document containing summary information on the goals of the initial 16 field trials, including the types of technology tested and industry focus. In June 2006, USDA released a preliminary progress report for these first 16 field trials but stated in the report that due to the timing of work plan submissions and the subsequent need for approved time extensions to complete proposed projects, 10 of these 16 projects awarded in fiscal year 2004 had not yet submitted final reports. In addition, USDA stated in this report that to fully understand the projects' results, interested parties should contact cooperative agreement project administrators to learn more about the projects' specific activities. However, the report did not include contact information. In May 2007, USDA released a third and final report on the results of the fiscal year 2004 field trials and descriptions of the fiscal year 2005 field trials for distribution to state, industry, and other stakeholders.[Footnote 20] USDA started sharing the results of premises registration and outreach cooperative agreements by publishing weekly premises registration statistics, by state, beginning in December 2006 (for state premises registration statistics, see app. VI, table 4). However, USDA has not formally shared any information about the strategies used by the individual state projects, nor has the agency communicated to stakeholders successful or unsuccessful approaches to registering premises. While USDA states that some results of cooperative agreements have been shared publicly at numerous stakeholder meetings, animal health officials and industry representatives told us that not enough information exists about the results of NAIS cooperative agreements, and that more sharing of results, best practices, and lessons learned is needed. For example, a researcher applying for fiscal year 2005 field trial funding told us it was difficult to determine whether previous NAIS cooperative agreements had included work similar to what the applicant was proposing. In addition, industry groups, state animal health officials, and experts told us that livestock markets would benefit from more information concerning the retrofitting of animal tracking equipment. USDA Plans to Improve Oversight and Give States with Greater Participation in NAIS Some Flexibility in Using Cooperative Agreement Funds: As we have previously reported, increasing oversight, linking funding to performance milestones, and altering flexibility are accountability mechanisms that can be used by agencies to encourage improved performance during an award period.[Footnote 21] In fiscal year 2007, USDA plans to increase oversight activities for some state cooperative agreements, on the basis of state premises registration levels at the time of the November 2006 announcement. For example, the 27 states that had 25 percent or less of their premises registered at the time of the cooperative agreement announcement and were eligible for more than $82,000 in awards would receive only 90 percent of reserved funding until a midyear review period. According to USDA, designated representatives will determine the success of cooperative agreements during this midyear review, largely on the basis of the goals stated in the cooperative agreements' approved work plans. The 23 states with greater than 25 percent of premises registered, or eligible for awards of less than $82,000, are eligible to receive 100 percent of cooperative agreement funding without a midyear review (see app. VI, table 5). In addition, all nonprofit industry organizations that receive cooperative agreement funding in fiscal year 2007 will be eligible for 50 percent of approved funds, with an additional 25 percent of funds released following each successful third-and fourth- quarter review of interim reports required by USDA. Moreover, USDA has linked funding to participation levels by providing for increased spending flexibility for some state recipients of fiscal year 2007 premises registration cooperative agreements. USDA believes that cooperative agreement funding may be more appropriately used by states with greater numbers of premises registered to support the animal ID and tracking components of NAIS. For example, the 14 states that have achieved greater than 25 percent of premises registered may spend up to 40 percent of funds on animal tracking infrastructure, such as to support NAIS's integration with preexisting disease eradication programs or to share in the cost of data collection equipment for livestock markets and dealers. The 18 states that have registered between 11 percent and 25 percent of premises may spend up to 30 percent of their funding on animal tracking infrastructure, while the 8 states that have registered between 6 percent and 10 percent of premises may spend up to 20 percent of funding on animal tracking infrastructure. Finally, those states that have registered less than 6 percent of premises are required to spend 100 percent of funds on outreach and premises registration. Total NAIS Program Costs Have Not Been Determined, but USDA Recently Announced Plans to Develop a Cost-benefit Analysis: USDA has not determined the program costs for NAIS but recently announced plans to hire a contractor to conduct a cost-benefit analysis for NAIS, in part to more precisely forecast the economic effects of the program. The Senate Appropriations Committee and the House of Representatives have raised concerns in recent years about how USDA has spent funds to develop and implement NAIS. Finally, the views of our expert panel members are mixed concerning NAIS's potential impact on the livestock industry. NAIS Costs and Benefits Are Not Known: Although implementation of NAIS began in 2004, USDA has not developed a comprehensive cost estimate or cost-benefit analysis for the program. In addition, to our knowledge, no industry group, academic institution, or state animal health agency has published a cost estimate for implementing and maintaining NAIS. Without a comprehensive cost-benefit analysis for NAIS, it is not known how much is required in federal, state, and industry resources to achieve rapid and effective traceback, or whether the potential benefits of the program outweigh the costs. In 2004, and again in 2006, the NAIS Subcommittee recommended that USDA prepare an in-depth, cost-benefit analysis for NAIS as part of the strategic planning process. In addition, 29 of the 32 expert panel members said that USDA should definitely or probably publish a cost- benefit analysis that contains detailed NAIS cost and benefit information for the different sectors of the livestock industry, states, and USDA. The NAIS working groups, other livestock industry representatives, and state animal health officials we interviewed said that the cost of implementing NAIS remained one of their biggest concerns. For example, in comments to USDA in 2005, the Livestock Marketing Association wrote that it is "highly critical of the fact that too little has been known" about the potential costs of establishing a national animal ID system and about who will bear those costs. Furthermore, the association wrote that a cost-benefit analysis is "long overdue" and that without better information, NAIS appears to be prohibitively expensive for the livestock industry to implement. As a result, without a reliable cost-benefit analysis that is consistent with federal guidance, stakeholders are unlikely to participate in NAIS due to their uncertainty that NAIS program benefits outweigh program costs. USDA announced plans in March 2007 to conduct a cost-benefit analysis for NAIS. USDA officials told us that the cost-benefit analysis will be used for program planning and resource allocation, producer and industry education, and public relations and outreach and to more precisely forecast the economic effects of NAIS. USDA officials anticipate that the cost-benefit analysis will be available in 2008. As we have previously reported, measuring the economic performance of federal programs, such as the extent to which program benefits exceed costs (net benefits) or are achieved at least cost (cost- effectiveness), could be a useful way to assess, in conjunction with other measures, the extent to which federal programs are meeting the nation's priorities.[Footnote 22] In addition, OMB has established general guidance on conducting cost-benefit analyses of federal programs to promote efficient resource allocation through well-informed decision making.[Footnote 23] OMB suggests that agencies follow this guidance in conducting analyses used to support government decisions to initiate, renew, or expand programs or projects that would result in a series of measurable benefits or costs extending 3 or more years into the future. The USDA announcement suggests that the planned NAIS cost- benefit analysis follow this and other available federal guidance. Concerns Exist over How USDA Has Spent Funds to Develop and Implement NAIS: In fiscal year 2004, the Secretary of Agriculture transferred $18.8 million from the Commodity Credit Corporation (CCC) to develop and implement NAIS, as shown in table 1. Although approximately $85.0 million had been made available for NAIS implementation by the end of fiscal year 2006, USDA had obligated only about $61.1 million as of late March 2007;[Footnote 24] thus, the agency has carried over about $23.9 million in unobligated NAIS funds into fiscal year 2007.[Footnote 25] Because NAIS funding has been designated by Congress to be available until expended, USDA can carry funds that it did not expend in prior years forward into the current year. In addition to these carryover funds, Congress appropriated an additional $33.0 million for the NAIS program for fiscal year 2007. The President's Budget requested $33.1 million for NAIS in fiscal year 2008. Table 1: USDA NAIS Budget Data, Fiscal Years 2004 (CCC Funds) through 2006: Dollars in thousands. Funding availability; Fiscal year: 2004: $18,793; Fiscal year: 2005: $33,197; Fiscal year: 2006: $33,007; Total: $84,997. Planned obligations: Information technology development, maintenance, and operations; Fiscal year: 2004: $2,009; Fiscal year: 2005: $6,858; Fiscal year: 2006: $7,733; Total: $16,600. Cooperative agreements; Fiscal year: 2004: 14,357; Fiscal year: 2005: 17,050; Fiscal year: 2006: 13,882; Total: 45,288. Communications and outreach; Fiscal year: 2004: 2,137; Fiscal year: 2005: 3,474; Fiscal year: 2006: 1,940; Total: 7,551. Headquarters and field staff and materials; Fiscal year: 2004: 290; Fiscal year: 2005: 3,125; Fiscal year: 2006: 5,285; Total: 8,700. Uncommitted/Unassigned funding; Fiscal year: 2004: 0; Fiscal year: 2005: 2,690; Fiscal year: 2006: 4,167; Total: 6,857. Total; Fiscal year: 2004: $18,793; Fiscal year: 2005: $33,197; Fiscal year: 2006: $33,007; Total: $84,997. Actual obligations: Information technology development, maintenance, and operations; Fiscal year: 2004: $1,829; Fiscal year: 2005: $5,276; Fiscal year: 2006: $2,466; Total: $9,571. Cooperative agreements; Fiscal year: 2004: 13,944; Fiscal year: 2005: 15,031; Fiscal year: 2006: 6,026; Total: 35,000. Communications and outreach; Fiscal year: 2004: 2,137; Fiscal year: 2005: 2,719; Fiscal year: 2006: 1,640; Total: 6,495. Headquarters and field staff and materials; Fiscal year: 2004: 379; Fiscal year: 2005: 3,213; Fiscal year: 2006: 6,428; Total: 10,019. Total; Fiscal year: 2004: $18,288; Fiscal year: 2005: $26,238; Fiscal year: 2006: $16,559; Total: $61,086. Unobligated funds; Fiscal year: 2004: $505; Fiscal year: 2005: $6,959; Fiscal year: 2006: $16,448; Total: $23,911. Actual expenditures: Information technology development, maintenance, and operations; Fiscal year: 2004: $1,813; Fiscal year: 2005: $3,946; Fiscal year: 2006: $908; Total: $6,668. Cooperative agreements; Fiscal year: 2004: 11,831; Fiscal year: 2005: 9,799; Fiscal year: 2006: 2,219; Total: 23,849. Communications and outreach; Fiscal year: 2004: 2,103; Fiscal year: 2005: 1,598; Fiscal year: 2006: 216; Total: 3,918. Headquarters and field staff and materials; Fiscal year: 2004: 379; Fiscal year: 2005: 3,174; Fiscal year: 2006: 6,427; Total: 9,979. Total; Fiscal year: 2004: $16,127; Fiscal year: 2005: $18,518; Fiscal year: 2006: $9,770; Total: $44,414. Source: USDA data as of March 27, 2007. Note: All figures have been rounded to the nearest thousand. [End of table] The Senate Appropriations Committee and the House of Representatives have raised concerns in recent years about how USDA has spent funds to develop and implement NAIS. For example, in the 109th Congress, the House passed a fiscal year 2007 appropriations bill for agriculture (H.R. 5384) that included a provision prohibiting funds from being obligated on NAIS until the House Appropriations Committee received a detailed plan for NAIS "including, but not limited to, proposed legislative changes, cost estimates, and means of program evaluation," and that the plan be published in the Federal Register for public comment. Although the bill, with the provision limiting obligations, passed in the House, it did not become law. USDA officials told us they have plans to obligate all carryover funds in fiscal year 2007. These plans include awarding additional cooperative agreements to states and industry organizations to register premises; additional investments in information technology development, maintenance, and operations; and communications and outreach. Expert Views Concerning Changes to the Livestock Industry Are Mixed: Questions have been raised about whether NAIS could lead to greater contracting, vertical integration, or horizontal consolidation in the livestock industry market structure,[Footnote 26] and whether NAIS could affect prices at both the retail and producer levels. Expert panel members provided the following views relating to changes in market structure as well as to changes in costs and prices for various market participants due to the implementation of NAIS. * The 32 experts were evenly split on whether contracting or horizontal consolidation would increase as a result of NAIS--16 said those effects would definitely or probably be more likely to occur, and 16 said those effects would not occur or are probably less likely to occur. * Twenty experts said vertical integration would not occur or is probably less likely to occur, while 12 said vertical integration is definitely or probably more likely to occur. * Regarding price effects at the retail level of meat and/or animal products, 12 experts thought prices would be higher; 15 thought there would be no effect; and 5 believed they would be lower. * When asked what would be the effect on prices paid to producers for livestock if NAIS led to increased costs in livestock markets and/or slaughter facilities, 21 experts believed that prices would probably or definitely decrease; 8 thought there would be no effect; and 2 thought that there would probably be an increase. * Similarly, when asked what would be the effect on prices paid to producers for livestock if NAIS led to decreased costs in livestock markets and/or slaughter facilities, 16 experts replied that prices paid to producers would probably or definitely increase; 13 replied that there would be no effect; and 1 respondent thought there would be a decrease. Conclusions: NAIS provides USDA, states, and the livestock industry with a historic opportunity for the United States to develop a comprehensive, coherent program to identify the nation's livestock animals and premises and achieve the goal of rapid and effective disease traceback. In addition, a successful program in sync with our international trading partners and competitors could boost consumer confidence in U.S. animal products and help maintain and expand market access. However, for NAIS to be fully effective and efficient in responding to an animal disease emergency, adequate levels of participation need to be achieved in all three NAIS components--premises registration, animal ID, and tracking. If insufficient numbers of animals are identified and tracked, the system will have gaps, despite millions of dollars being invested in the program's development. Conversely, high levels of participation would better position the United States to handle future animal health emergencies and minimize economic, trade, and possibly human health consequences. USDA's changes in direction over the past 3 years have caused considerable confusion and frustration among many NAIS stakeholders, and the program's implementation may be in danger of losing momentum. Most critically, whether NAIS is voluntary or mandatory, the lack of participation benchmarks makes it more difficult to gauge progress in attaining the necessary levels of participation for an effective animal ID program and, if there is insufficient participation, to develop strategies to achieve it. Without meaningful progress, USDA's expenditures on NAIS will continue to be questioned. In addition, if USDA does not resolve several key implementation issues, the program will continue to face opposition by some industry stakeholders, and participation in all three NAIS components could be limited. Collectively, these unresolved issues will likely lead to ineffective and inefficient implementation and prevent NAIS from achieving the goal of rapid and effective traceback. First, prioritizing how NAIS is implemented, such as by species, would allow USDA and stakeholders to better allocate their resources and improve the program over time. Second, the integration of NAIS with other USDA and state animal disease eradication programs and branding systems would remove another hurdle preventing participation in NAIS's animal ID and tracking components. Third, creating a robust process for selecting, setting standards for, and independently testing and evaluating animal ID and tracking devices in meeting NAIS standards is important to ensure effectiveness and interoperability across the national program and, therefore, would encourage investment. Fourth, identifying time-sensitive and cost-effective goals for traceback, which may vary by disease, would allow stakeholders to have common goals in responding to an event, potentially speeding up response and, therefore, minimizing economic losses. Fifth, requiring the recording of information in NAIS databases that may be critical for efficient traceback, such as species, approximate age, or date of birth, would enable animal health authorities to more quickly locate only those premises and animals that are relevant in an investigation, thus minimizing time and resources and hastening response. Finally, we believe that if USDA were to provide industry, state, and other stakeholders with key information on the results of cooperative agreements, it would help identify the most effective and efficient means to implement the program and likely increase participation and enable producers, livestock markets, states, and other stakeholders to make informed decisions about where to allocate scarce resources. Moreover, for planning purposes in allocating federal, state, and industry resources, it is important for stakeholders, Congress, and the public to know how much it will cost to implement and maintain NAIS, compared with its benefits. Without a reliable cost-benefit analysis, stakeholders are unlikely to participate in NAIS due to their uncertainty about whether program benefits outweigh the costs. Recommendations for Executive Action: To achieve the program's goal of rapid and effective animal disease traceback, we recommend that the Secretary of Agriculture direct the Administrator of APHIS to reestablish participation benchmarks to gauge progress in registering premises and identifying and tracking animals; monitor participation; and, if participation does not meet the benchmarks, take further action, such as making participation mandatory or creating incentives to achieve those levels of participation. In addition, we recommend that the Secretary direct the Administrator of APHIS to take the following seven actions to implement NAIS more effectively and efficiently and achieve the program's goal of rapid and effective traceback: * set priorities, in consultation with the NAIS species working groups, state animal health officials, and others, for implementing NAIS incrementally by species or other criteria; * determine how NAIS will integrate with existing USDA and state animal disease eradication programs and branding systems; * establish a robust process to select, standardize, and independently test and evaluate the performance of animal ID and tracking devices to ensure they meet minimum standards; * identify--in consultation with the NAIS species working groups, state animal health officials, and others--current baselines for animal disease traceback, and develop time-sensitive, cost-effective goals for traceback under NAIS, which may include separate time frames for specific diseases; * evaluate what information is critical for efficient traceback, such as species, approximate age or date of birth, and require that participants record that information in the NAIS animal ID and tracking databases; * increase the monitoring of NAIS cooperative agreements, and evaluate and publish the results of cooperative agreements on a timely basis; and: * conduct the planned analysis of the costs and benefits of NAIS following criteria established in OMB guidance for conducting cost- benefit analyses for federal programs and publish the results for comment. Agency Comments and Our Evaluation: We provided a draft of this report to USDA for review and comment. In written comments on our draft report, USDA stated that it appreciated our comprehensive evaluation of NAIS and generally agreed with our recommendations. However, regarding our recommendation that USDA establish a robust process to select, standardize, and independently test and evaluate the performance of animal ID and tracking devices to ensure they meet minimum standards, USDA agreed with the need to establish a more robust process for having ID devices tested to meet minimum performance standards, but believed that these standards must be defined through a consensus of affected stakeholders. USDA stated that as performance standards are established, the selection of such devices will then be warranted. USDA also stated that as part of the evaluation process, it will specify the testing standards and then review the manufacturer's documentation of laboratory testing and field trials. In addition, USDA stated that testing of such devices should be at the expense of the device manufacturer. We recognize the need for USDA to work with stakeholders before determining which ID and tracking devices are most appropriate for NAIS. However, the sooner USDA selects specific technologies, the sooner producers, livestock markets, slaughter facilities, and others will likely participate in the animal ID and tracking components of NAIS. As a starting point, some NAIS working groups have recommended specific ID devices for their species. It is common U.S. practice to select one technology for systems that need to be widely implemented in different environments, and we have previously reported that a robust process for selecting, standardizing, and testing and evaluating technologies leads to the most effective and efficient systems. During the course of our work, we found that USDA's technology-neutral position has caused producers, market operators, and slaughter facilities to be generally discouraged from investing in new animal ID or tracking devices for NAIS due to fear that their choices might be inconsistent with others in the marketplace, or that USDA might adopt specific devices in the future. The selection of specific ID and tracking devices, therefore, would ensure consistency and interoperability across the program's many potential users, leading to more efficient implementation. Furthermore, as stated in this report, USDA has recognized the need for animal ID technologies that are compatible with Canada and Mexico, and it also has stated that harmonizing the United States' program with other countries will facilitate safe trade. Selecting technologies for NAIS that are in sync with our trading partners and competitors could have positive trade implications for the United States. Lastly, the selection of specific devices would inform the cost-benefit analysis that USDA is currently conducting, which, in turn, could affect participation levels. These reasons underscore the need for USDA to select specific ID and tracking devices, on the basis of independent, reliable information regarding their performance in meeting minimum standards and of consultations with stakeholders. In its written comments, USDA also provided points of clarification and provided details about current and future actions that the agency plans to take to address our recommendations, which we incorporated throughout the report, as appropriate. USDA's written comments and our specific responses appear in appendix VII. In addition, USDA provided technical comments that we incorporated throughout the report, as appropriate. As we agreed with your office, unless you publicly announce the contents of this report earlier, we plan no further distribution of it until 30 days from the date of this report. At that time, we will send copies of this report to the interested congressional committees, the Secretary of Agriculture, and other interested parties. We will also make copies available to others upon request. In addition, this report will be available at no charge on the GAO Web site at http://www.gao.gov. If your or your staff have any questions about this report, please contact me at (202) 512-3841 or ShamesL@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix VIII. Sincerely yours, Signed by: Lisa Shames: Director, Natural Resources and Environment: [End of section] Appendix I: Objectives, Scope, and Methodology: The objectives of our review were to determine (1) how effectively the U.S. Department of Agriculture (USDA) is implementing the National Animal Identification System (NAIS) and, specifically, the key implementation issues identified by livestock industry groups, market operators, state animal health officials, and others; (2) how USDA has distributed cooperative agreement funding to help states and industry prepare for NAIS and evaluated the agreements' results; and (3) what USDA and others estimate are the costs for USDA, states, and the livestock industry to implement and maintain NAIS. To address these objectives, we interviewed USDA officials responsible for implementing NAIS and conducted site visits to selected livestock markets and cooperative agreement field trials. We conducted structured interviews in person or via telephone with animal health officials in seven states: California, Iowa, Michigan, New Mexico, Texas, Vermont, and Wisconsin. These states were selected on the basis of their geographic dispersion; the range in the number of premises located in each state; and, in some cases, their high levels of livestock production. We also conducted interviews in person or via telephone and reviewed documents from the NAIS Subcommittee and the 10 NAIS working groups that report to the NAIS Subcommittee: Beef and Dairy Cattle Working Group, Bison Working Group, Camelid Working Group, Cervid Working Group, Equine Working Group, Goat Working Group, Market/ Processor Working Group, Poultry Working Group, Sheep Working Group, and Swine Working Group. We also conducted structured interviews in person or via telephone and reviewed documents from four industry organizations: American Farm Bureau Federation, Livestock Marketing Association, National Cattlemen's Beef Association, and National Livestock Producers Association. In addition, we reviewed documents from the United States Animal Health Association Livestock ID Committee; the Ranchers-Cattlemen Action Legal Fund --United Stockgrowers of America; three NAIS opposition groups--NoNAIS.org, Liberty Ark Coalition, and the Farm and Ranch Freedom Alliance; and other organizations that testified before Congress on NAIS in recent years or spoke at USDA's listening sessions in 2004. We attended the NAIS Subcommittee meeting and a USDA meeting with state departments of agriculture on NAIS held in August 2006 in Kansas City, Missouri, and the Secretary's Advisory Committee on Foreign Animal and Poultry Diseases meeting in September 2006 in Riverdale, Maryland. We identified and reviewed applicable laws, USDA policies, guidance, and technical standards regarding NAIS. We also reviewed relevant GAO reports and a Congressional Research Service report to Congress on animal identification (ID) and traceability. To determine how USDA has distributed cooperative agreement funding to help states and industry prepare for NAIS, we reviewed USDA documentation related to cooperative agreements signed between USDA and states, territories, tribes, and industry groups from fiscal years 2004 through 2007. To determine which livestock species were the focus of cooperative agreement field trials, we reviewed and systematically recorded this information from cooperative agreement recipients' work plans; however, we did not independently assess whether the proposed work with these species and industry sectors took place. To determine how USDA has evaluated the results of cooperative agreements, we interviewed NAIS program staff, reviewed guidance provided to recipients, and reviewed quarterly and final reports submitted to USDA by cooperative agreement recipients. Because some cooperative agreements were ongoing and because other recipients did not report to USDA in a timely manner, we could not examine a complete set of quarterly and final reports for all recipients. We conducted a reliability assessment of the data that USDA provided to us on the NAIS cooperative agreements and found these data to be reliable for our reporting purposes. For the third objective, to determine estimates of the costs to implement and maintain NAIS, we asked representatives from USDA, industry groups, academic institutions, and state animal health agencies for any NAIS cost estimates they had developed. We identified and reviewed federal guidance for developing cost estimates and cost- benefit analyses and sound economic and cost accounting principles. We also reviewed NAIS budget data from USDA for fiscal years 2004 through 2007, and conducted a reliability assessment of these data and found them to be reliable for our reporting purposes. To help answer the first and third objectives, we convened a Web-based panel of 32 experts on several aspects of NAIS. The process we followed is based on GAO guidance for identifying experts for panels or other work requiring expertise in a specific area. We identified potential panel members by conducting a literature search to obtain the names of individuals who had published on animal ID in academic journals and in other relevant venues. We also asked for recommendations from individuals we interviewed for other aspects of the job. We then selected individuals who were actively involved in the development or implementation of NAIS and were knowledgeable of its details; who had conducted research, or were published in peer-reviewed journals on animal ID; or who were recognized by their peers as an expert on NAIS. Panel members were asked to fill out a Web-based questionnaire, which asked for their beliefs and opinions on future participation in NAIS, effective traceback, implementation of NAIS, the costs and benefits of NAIS, the impact of NAIS on the livestock industry and consumers, ID technology, and databases for tracking animals. Panel members had approximately 3 weeks to fill out their questionnaires in December 2006 and January 2007. All panel members completed their questionnaires, giving us a 100 percent response rate. The questions and aggregated responses are presented in appendix IV. While we display only the quantitative, closed-ended responses, we also relied on the responses to the qualitative, open-ended questions to inform our findings in this report. The views expressed by the panel members do not necessarily represent the views of GAO. We conducted our work from June 2006 to May 2007 in accordance with generally accepted government auditing standards. [End of section] Appendix II: Select Domestic and Foreign Animal Diseases of Concern Identified by USDA: Table: Domestic: Animal disease: Avian influenza (low pathogenic); Livestock animals affected: Poultry; Can affect humans?: No. Animal disease: Bovine spongiform encephalopathy; Livestock animals affected: Cattle; Can affect humans?: Yes. Animal disease: Bovine brucellosis and Swine brucellosis; Livestock animals affected: Cattle, bison, goats, swine, and cervids; Can affect humans?: Yes. Animal disease: Chronic wasting disease; Livestock animals affected: Cervids; Can affect humans?: Unknown. Animal disease: Equine infectious anemia; Livestock animals affected: Horses, donkeys, mules, ponies, and zebra; Can affect humans?: No. Animal disease: Johne's disease; Livestock animals affected: Cattle, sheep, goats, and cervids; Can affect humans?: Unknown. Animal disease: Pseudorabies; Livestock animals affected: Swine, cattle, sheep, and goats; Can affect humans?: No. Animal disease: Scrapie; Livestock animals affected: Sheep and goats; Can affect humans?: Unknown. Animal disease: Texas (splenetic) fever; Livestock animals affected: Cattle; Can affect humans?: No. Animal disease: Tuberculosis; Livestock animals affected: Cattle, bison, and captive cervids; Can affect humans?: Yes. Foreign: Animal disease: Avian influenza (highly pathogenic); Livestock animals affected: Poultry; Can affect humans?: Yes. Animal disease: Dourine; Livestock animals affected: Horses and donkeys; Can affect humans?: No. Animal disease: Exotic Newcastle disease; Livestock animals affected: Poultry; Can affect humans?: Yes. Animal disease: African swine fever; Livestock animals affected: Swine; Can affect humans?: No. Animal disease: Classical swine fever; Livestock animals affected: Swine; Can affect humans?: No. Animal disease: Contagious bovine pleuropneumonia; Livestock animals affected: Cattle and bison; Can affect humans?: No. Animal disease: Foot-and-mouth disease; Livestock animals affected: Cattle, sheep, goats, and swine; Can affect humans?: Yes, but rarely infects humans. Animal disease: Glanders; Livestock animals affected: Horses, donkeys, mules, and goats; Can affect humans?: Yes. Source: USDA. Note: Pursuant to the Public Health Security and Bioterrorism Preparedness and Response Act of 2002 (the "Bioterrorism Act of 2002"), USDA identifies animal diseases that have the potential to pose a severe threat to livestock and human health, including bacillus anthracis, brucella abortusm, brucella melitensis, brucella suis, burkholderia mallei, burkholderia pseudomallei, clostridium botulinum, coccidioides immitis, francisella tularensis, botulinum neurotoxins, clostridium perfringens epsilon toxin, shigatoxin, staphylococcal enterotoxins, T-2 toxin, African horsesickness, peste des petits ruminants, swine vesicular disease virus, lumpyskin disease virus, sheep pox, and goat pox. USDA identifies other domestic animal diseases of concern in federal regulations, including acute swine erysipelas (affects swine), anthrax (all domestic species), bluetongue (all domestic species), chlamydiosis (poultry), Eastern equine encephalomyelistis (horses), mycoplasma gallisepticum (poultry), mycoplasma meleagridis (poultry), mycoplasma synoviae (poultry), salmonella enteritidis (poultry), salmonella gallinarum (poultry), salmonella pullorum (poultry), scabies (cattle), infectious salmon anemia, and spring viremia of carp. Additional foreign animal diseases of concern identified by USDA include equine viral arteritis (horses, donkeys, mules, ponies, and zebra); hendra (horses); nipah (swine and horses); Rift Valley fever (cattle, sheep, and goats); rinderpest (cattle, sheep, and goats); Venezuelan equine encephalomyelitis (all equine species); and vesicular stomatitis (swine, cattle, sheep, and goats). [End of table] [End of section] Appendix III: Members of GAO's Expert Panel on NAIS: This appendix provides the names and affiliations of 32 academic, government, and other experts who, as members of our expert panel on NAIS, completed a Web-based questionnaire from December 2006 to January 2007 regarding USDA's implementation of NAIS. We also spoke with a select number of these experts regarding animal ID and tracking technology, among other issues. Expert Panel Members: * Dr. David P. Anderson, Associate Professor and Extension Economist - Livestock and Food Products Marketing, Department of Agricultural Economics, Texas A&M University: * Dr. DeeVon Bailey, Interim Department Head and Professor, Department of Economics, Utah State University: * Dr. Joseph Balagtas, Assistant Professor, Department of Agricultural Economics, Purdue University: * Dr. Dale A. Blasi, Professor & Extension Beef Specialist, Department of Animal Sciences & Industry, Kansas State University: * Dr. D. Scott Brown, Research Assistant Professor and Program Director of Livestock and Dairy, Food and Agricultural Policy Research Institute, University of Missouri: * Dr. Daniel D. Buskirk, Associate Professor and Beef Extension Specialist, Department of Animal Science, Michigan State University: * Dr. Julie A. Caswell, Professor and Department Chair, Department of Resource Economics, College of Natural Resources and the Environment, University of Massachusetts: * Dr. David A. Daley, Professor, College of Agriculture, California State University, Chico: * Dr. Kevin Dhuyvetter, Agricultural Economist, Department of Agricultural Economics, Kansas State University: * Dr. Basil Eastwood, National Program Leader, Plant and Animal Systems, Cooperative State Research, Education, and Extension Service, USDA: * Dr. Scott Greiner, Associate Professor and Extension Animal Scientist, Beef and Sheep, Department of Animal & Poultry Sciences, College of Agriculture and Life Sciences, Virginia Polytechnic Institute and State University: * Dr. Ron A. Gustafson, Senior Economist, Beef Analysis, Economic Research Service, USDA: * Dr. James C. Heird, Director of Equine Sciences Department, Equine Teaching and Research Center, Colorado State University: * Dr. Julie Jarvinen, Associate Professor, Department of Veterinary Pathology, Iowa State University: * Dr. Cleon V. Kimberling (retired), Clinical Sciences Department, Colorado State University: * Dr. John D. Lawrence, Professor and Extension Livestock Economist, Agricultural Economics, Department of Economics, Iowa State University: * Dr. Darrell R. Mark, Assistant Professor and Livestock Extentionist, Department of Agricultural Economics, University of Nebraska, Lincoln: * Dr. Bret D. Marsh, State Veterinarian, Indiana State Board of Animal Health, and immediate past president of the United States Animal Health Association: * Dr. James D. McKean, Extension Veterinarian and University Professor, Department of Veterinary Diagnostic & Production Animal Medicine, College of Veterinary Medicine, Iowa State University: * Mr. Douglas O'Brien, Co-Director, National Agricultural Law Center, University of Arkansas School of Law, and Staff Attorney, Drake University Agricultural Law Center: * Dr. James W. Oltjen, Professor and Extension Specialist, Animal Management Systems, Department of Animal Science, University of California, Davis: * Dr. Derrell Peel, Professor, and Livestock Extensionist, Department of Agricultural Economics, Oklahoma State University: * Dr. Valerie Ragan, President, AgWorks Solutions LLC, and former Assistant Deputy Administrator for USDA's Animal and Plant Health Inspection Service, Veterinary Services: * Dr. Kris Ringwall, Animal Scientist and Director of the Dickinson Research Extension Center, North Dakota State University, and Executive Secretary, North Dakota Beef Cattle Improvement Association: * Dr. Joan Dean Rowe, Associate Professor, Department of Population Health & Reproduction, University of California, Davis: * Dr. Ted Schroeder, Professor and Director of Graduate Program, Department of Agricultural Economics, Kansas State University: * Dr. Clifford F. Shipley, Clinical Associate Professor, College of Veterinary Medicine, University of Illinois: * Dr. Ronnie E. Silcox, Associate Professor and Extension Beef Specialist, Animal & Dairy Science Department, University of Georgia: * Dr. Michael A. Tomaszewski, Professor and Extension Dairy Specialist, Department of Animal Science, Texas A&M University: * Dr. Glynn Tonsor, Assistant Professor, Department of Agricultural Economics, Michigan State University: * Dr. Wendy J. Umberger, Assistant Professor and Extension Economist, Department of Agricultural and Resource Economics, Colorado State University: * Dr. Kelly Zering, Associate Professor, Department of Agricultural and Resource Economics, North Carolina State University, Raleigh: [End of section] Appendix IV: GAO Expert Panel Questions and Responses on NAIS: Expert Panel: USDA's Implementation of the National Animal Identification System: We conducted the following survey as part of our review of USDA's implementation of NAIS. We received a 100 percent response rate from a panel of 32 experts who filled out a Web-based questionnaire in late December 2006 and early January 2007. For presentation purposes in this appendix, we have combined the category "No expertise on topic" with "No answer" and the category "50% or less" with "51 to 60%." However, when the experts filled out the questionnaire, those categories were not combined. The views expressed by the panel members do not necessarily represent the views of GAO. Part I: Participation in the NAIS Voluntary Program: In USDA's November 2006 draft National Animal Identification System (NAIS): A User Guide and Additional Information Resources (NAIS User Guide), USDA states that NAIS is a voluntary program that helps producers and animal health officials respond rapidly and effectively to animal disease events for livestock and poultry in the United States. The next three questions ask for your opinion on likely participation levels for the three components of NAIS (premises registration, animal identification, and animal tracking) under a voluntary program. In this questionnaire, the term "producer" refers to all individuals engaged in the ownership, management, or marketing of any of the species of livestock included in NAIS. For example, in the beef cattle industry, this refers to cow-calf producers as well as stocker, backgrounder, and feedlot operators. While owners or managers of certain species, for example horses, may not typically be referred to as producers, these individuals are included in this definition of producer. This definition is consistent with USDA's draft NAIS User Guide (pg. 5). The term "livestock market" refers to livestock auction markets, sale barns, and sale yards. Q1. What do you believe will be the percentage of premises registered for each sector of livestock production under the NAIS voluntary program? a. Producers; 60% or less: 24; 61-70%: 4; 71-80%: 2; 81-90%: 1; 91- 100%: 0; No expertise/ No answer: 1; Number of respondents: 32. b. Livestock Markets; 60% or less: 9; 61-70%: 4; 71-80%: 5; 81-90%: 6; 91-100%: 5; No expertise/ No answer: 3; Number of respondents: 32. c. Slaughter Facilities; 60% or less: 6; 61-70%: 1; 71-80%: 3; 81- 90%: 12; 91-100%: 8; No expertise/ No answer: 2; Number of respondents: 32. [End of table] Q2. What do you believe will be the percentage of animals identified (as individuals or, where applicable, as a group) for each species of animal under the NAIS voluntary program? a. Bison; 60% or less: 18; 61-70%: 1; 71-80%: 1; 81-90%: 0; 91-100%: 1; No expertise/ No answer: 11; Number of respondents: 32. b. Camelids (e.g., alpacas, llamas); 60% or less: 14; 61-70%: 1; 71- 80%: 2; 81-90%: 1; 91-100%: 0; No expertise/ No answer: 14; Number of respondents: 32. c. Cattle (beef); 60% or less: 24; 61-70%: 2; 71-80%: 2; 81-90%: 1; 91-100%: 0; No expertise/ No answer: 3; Number of respondents: 32. d. Cattle (dairy); 60% or less: 10; 61-70%: 4; 71-80%: 7; 81-90%: 5; 91-100%: 3; No expertise/ No answer: 3; Number of respondents: 32. e. Cervids (e.g., deer, elk); 60% or less: 14; 61-70%: 3; 71-80%: 0; 81-90%: 1; 91-100%: 1; No expertise/ No answer: 13; Number of respondents: 32. f