This report assesses the impact of IFPRI’s Global Research Program on The Sustainable Development of Less-Favored Areas (“GRP-5”). Initiated in 1998, the stated objectives of the research program were (a) to provide empirical evidence on appropriate development strategies and public investments for improving the well-being of individuals living in less-favored areas (LFAs); and (b) to assess the appropriate targeting of various public investments to favored versus less-favored areas. The program’s research activities generally were confined to addressing the first of these objectives. The GRP-5 research was primarily undertaken in Ethiopia, Honduras, and Uganda, using quantitative livelihoods and bio-economic modeling approaches to studying constraints and opportunities for poor households in less-favored areas (LFAs). In the first section of the report, we place this research program in the context of the body of work conducted within the CGIAR that has investigated the appropriate allocation of various public investments between favored and less-favored agroecological zones. The second section of the report provides a brief overview of the program’s research activities within each of the three countries of emphasis, along with the various research outputs. These research activities extended work on resource degradation and land management that IFPRI had been involved in prior to the initiation of GRP-5. Major workshops held in each country were the principal venues for dissemination of the research findings. A primary goal of these workshops was to influence individuals in positions of authority to act upon those findings, either in terms of instituting formal policies or programs, or fostering follow-up research more directly geared to implementation. The workshops were, however, by no means the only outputs from the program. Besides the workshop papers a wide range of publications was generated by the program including dissemination briefs, research reports, papers in journals (including special editions containing a series of papers generated by the project and related research), and a book published in 2006 entitled Strategies for Sustainable Land Management in the East African Highlands (Pender, Place, and Ehui 2006a). The third section of the report briefly reviews the extent to which the GRP-5 research program achieved its stated objectives. In Honduras, the operational approach concentrated exclusively on LFAs and was therefore incapable of addressing the basic issue of the appropriate allocation of resources and development effort between favored and less-favored areas. The research in both Uganda and Ethiopia did include areas of both high and low agricultural potential, and produced some results comparing the impacts of similar interventions in different agroecological domains; but the primary emphasis remained on LFAs. In part, this was no doubt related to financial constraints limiting the geographic extent of the projects’ fieldwork. But additionally, it may well reflect the practical difficulties of reconciling research themes of general interest within the CGIAR and the broader donor community (i.e., geographic allocation of research and investment funds) with exigencies of engaging local policymakers whose interest lay in understanding the opportunities and constraints conditioning the appropriate development strategies for different types of less-favored lands. The report’s fourth section discusses the study team’s findings, based on field visits to Ethiopia, Honduras, and Uganda, regarding perceptions of the influence and impact of the GRP-5 research activities. There was general agreement that IFPRI’s approach to the research was rigorous, well-conceived and well-executed, and that the information generated is highly useful as a description of the realities of agricultural households in LFAs. IFPRI researchers were uniformly praised for the care with which data collection efforts were undertaken and the rigor with which those data were analyzed. The training aspect of the research was generally acknowledged by those involved in the programs. This included both the formal graduate training and interaction with students and faculty staff at local academic institutions. Finally, there was widespread sentiment that the research had succeeded in drawing attention to, and contributing to policy debates surrounding, poverty issues in LFAs. IFPRI’s research is widely held to have established important baseline information for use in monitoring changes that may occur if and when policy initiatives are undertaken in the future. Some shortcomings were identified as well. Concerns about analytical methods were expressed by some (primarily non-economists). Others indicated that there was insufficient follow-up for the wider policy community or the general public after the high-profile summary workshops that presented the research findings, and that the academic nature of the research outputs was not directly relevant to policymakers. Finally, significant concern was expressed about substantial lags-upwards of three years-between the time the research was completed and the time the main research reports were published. The final sections of the report discuss tangible indications of impact on policy in the countries of emphasis. The GRP-5 work’s primary contribution was as a benchmark. There is a widespread sentiment among those with experience of the GRP-5 research program in the participating countries that it generated a useful and much-needed description of the socioeconomic conditions within which poor households operate in less-advantaged regions. This information has been of value in subsequent follow-on research in terms of problem definition, research focus, and (in some cases) site selection. It has also been useful in the design and implementation of some rural development projects as well. There was considerable variation in the extent to which the research program had a direct impact on policy or related programs. In the case of Uganda, the preliminary findings of IFPRI’s research were, at least for a time, fairly closely linked to the government’s agricultural priority-setting process. In the case of Ethiopia, in the early years of the research work there was a close relationship between the research team, a local university and the regional Bureau of Agriculture. As a result, research findings did inform state level policies and programs. But, as these relationships weakened over time, and particularly once the research work was completed, this influence waned. In Honduras, the high degree of collaboration between IFPRI researchers and their PRONADERS partners appeared to have set the stage for translating research results seamlessly into government policy. Unfortunately, the change in government midway through the data collection phase of the project altered the situation irrevocably. A number of factors are identified as contributing to the difficulty of translating the research findings into actionable policies and policy outcomes. First, there are several different audiences for research of the sort reviewed here, including the broader research and donor communities, the in-country policy community, and field practitioners. The information demands for each group are by no means the same, and in some cases there may in fact be little overlap. Second, the intellectual culture at IFPRI favoring academic research suitable for publication in scholarly journals can limit the relevance of the research to policy makers (and also to field practitioners). Third, limited on-site representation significantly restricts IFPRI’s ability to influence policy debates. In summary, the assessment team was left with the conclusion that the sort of research conducted under GRP-5 has significant potential usefulness to other follow-on research, as well as for the design of projects aimed at improving the well-being of smallholders in LFAs. But, it is far more difficult to see clear links to policymakers who approach their jobs with their own particular agendas. This in no way diminishes the value of the research per se, but it certainly calls into question its sustained influence on the policymaking process.