||In 1990, economists of the Southern Africa Development Community/International Crops Research Institute for the Semi-Arid Tropics (SADC/ICRISAT) and the International Maize and Wheat Improvement Center (CIMMYT) launched a policy analysis training and networking activity. The Policy Analysis Initiative (PAI) offers analytical training and advisory assistance on issues of cereal grain policy analysis to economists based in national agricultural research systems (NARS) in eastern and southern Africa. The PAI seeks to improve the quality and relevance of economic analysis conducted by these economists. In addition, PAI facilitates communication among economists about common research themes and analytical methods. The work reported in this volume was preceded by two workshops. The first was held at Matopos Research Station, Bulawayo, Zimbabwe from 26-28 September 1990. The three-day workshop provided the opportunity for participants in the PAl to gather together to discuss their research objectives, hypotheses and methods. The resource people reviewed common themes cutting across the proposed analyses. Common objectives, hypotheses, methods of analyses and data source were reviewed for: fertilizer and other agricultural input policy studies; grain market policy studies, and research resource allocation studies. Plans for interaction over the next 12 months especially with resource people, were discussed. In the second workshop, held at Matopos from 16-20 March 1992, 15 PAl participants reported the progress of their work in a set of formal conference papers. The participants commented on each others' work. Resource people commented on each participant's presentation, and later followed this with detailed written comments. Two days were spent in a fertilizer policy seminar conducted by Dr. Vasant Gandhi of International Food Policy Research Institute (lFPRI), who lectured, presented literature, and organized a hands-on computer analysis session. The three resource people from SADC/ICRISAT and CIMMYT also spent a considerable amount of time assisting the participants on an individual basis both before and after the reporting workshop. The papers included in this volume are a result of research undertaken by the economists participating in the PAl. They are divided into four categories. Section I of this volume has six papers that address the issues of technology adoption. Section II presents three papers dealing with fertilizer policy. The four papers of section III analyze grain marketing issues. Section IV contains two papers that focus on research priority setting. Economists working in NARS in eastern and southern Africa have generally not exploited their comparative advantage in contributing to agricultural policy formulation. This advantage lies in their understanding of how policies are implemented in rural areas and how they affect the decisions of specific groups of farmers. This view of the links between macro- and micro-level decision-making relates most directly to the process of technology generation and adoption. But NARS economists are also positioned to contribute to the broader range of discussions about market policy and agricultural development strategy. The 15 papers in this volume begin to show these contributions.