This is the fourth Annual Report on Results and Impact of IFAD Operations (ARRI) produced by the Office of Evaluation (OE). As in the past, the ARRI report consolidates and synthesizes the results and impact of IFAD's operations based on a cohort of project and country programme evaluations conducted in 2005. It presents an analysis of last year's evaluation findings and a comparison with results of previous years (2002-2004), with the objectives of contributing to accountability in terms of IFAD's performance, and of learning from evaluation findings. The ARRI report's key findings aim to prompt discussion of necessary corrective action and changes in the way IFAD conducts its business. The ARRI report does not focus on the follow-up action taken by the Programme Management Department to evaluation recommendations, as these are discussed in the President's Report on the Implementation Status of Evaluation Recommendations and Management Action.

This year's ARRI report introduces target scores for each evaluation criterion as an internal benchmark to illustrate how such scores can be used for performance and results  management. This is an experimental exercise to demonstrate the usefulness of such a system rather than to suggest a specific target score. Performance targets will focus on how effectively IFAD is addressing the criteria that are important to development effectiveness. The assumptions used in this ARRI report are based on a six-point rating scale that introduces greater differentiation between ratings and allows for a more nuanced performance assessment 1. Using this scale, it is reasonable, on balance, to expect 5 per cent of projects to be highly successful and 50 per cent to be successful. If an organization cannot achieve a success rate of more than 50 per cent, questions about its performance may be justified. It is also true that given the difficult and innovative nature of IFAD's work, some relative shortcomings are understandable. Therefore, targets of 20 per cent of projects being classed as moderately successful, 15 per cent as moderately unsuccessful and 5 per cent being classed in the bottom two categories (unsuccessful and highly unsuccessful) might be expected. These percentages produce a target mean of 4.2, which is just above a moderately successful rating. The only exception is relevance (one of the three project performance criteria), for which a mean target rating of 5 was set, as clearly all IFAD projects should have this characteristic.

The ARRI report follows the same structure as previous years. Section II provides an outline of the evaluations conducted in 2005. Sections III-V provide a synthesis of evaluation findings for project performance, rural poverty impact and partner performance, which are the main evaluation criteria used by IFAD (annex I provides an overview of the methodological framework). Section VI summarizes overall achievements and discusses the contribution of evaluated projects to IFAD's strategic objectives and to the Millennium Development Goals (MDGs). Finally, section VII presents the report's key findings and recommendations.