Agriculture accounts for about 30 percent of Africa’s gross domestic product (GDP) and 75 percent of total employment (World Bank 2007). However, nearly half of the area of Africa, which is home to more than 14 percent of the low-income countries in the world, is either arid or semi-arid, and over 90 percent of agricultural production is rain-fed (Fisher et al. 2004; WDI 2005). This implies that erratic rainfall patterns present serious challenges to food production in these areas (Fisher et al. 2004), and this will be further worsened by climate change which is expected to increase rainfall variability in many African countries that are already at least partly semi-arid and arid. These concerns are substantial in Ethiopia where the agriculture sector—the most important sector for poverty reduction—has been undermined by lack of adequate plant-nutrient supply, depletion of soil organic matter, and soil erosion (Grepperud 1996). In an effort to overcome these challenges, the government and non-governmental organizations have consistently promoted chemical fertilizer as a yield-augmenting technology. Despite this promotion, chemical fertilizer adoption rates remain very low (Byerlee et al. 2007, 37), and in some cases, there is evidence suggesting a retreat from fertilizer adoption (EEA/EEPRI 2006), possibly due to escalating fertilizer prices and production and consumption risks (Kassie, Yesuf, and Köhlin 2008; Dercon and Christiaensen 2007). More importantly, government policies to promote technologies lack a clear understanding of the role of agroecology, such as rainfall, in conditioning the effectiveness of technologies in enhancing productivity. The distribution and amount of rainfall varies both in spatial and temporal terms across and within Ethiopia