As noted by Duflo et al. , the rapid population growth has made Africa to be no longer viewed as a land-abundant region where food crop supply could be increased by expansion of land used in agriculture. Large areas in Africa are increasingly becoming marginal for agriculture and arable land has become scarce in many African countries. This makes the need for intensification of land use through use of productivity enhancing technologies such as fertilizer critical for achieving food security. Yet, the rate of increase in fertilizer use has been substantially lower in Africa than in Asia and Latin America . Similar observations are also made by Ariga et al., . According to Howard et al., , high external input technologies, lack of infrastructure, research, development, and even extension are major obstacles to increasing fertilizer application rates in sub-Saharan Africa. The fertilizer supply is limited and the cost is prohibitive for SSA farmers because fertilizer may cost as much as five times the global market price . This problem has been aggravated by the recent spike in world fertilizer prices. For instance, after accounting for inland transport costs, the wholesale price of DAP fertilizer in Nakuru, Kenya, rose from 1,750 Ksh per 50kg bag in 2007 to nearly 4,000 Kshper 50kg bag in 2008 . Consequently,pot with drainage holes fertilizer application rates in SSA have remained the lowest in the world and continue to decline even though soils in SSA are considered as poor as those in Latin America and Asia .

Despite the low application rates of fertilizer in SSA relative to other parts of the world, studies have shown that Kenya’s fertilizer use relative to those of the countries in the region has increased dramatically since its fertilizer market was liberalized in the early 1990s. Kenya is the only country in Sub-Saharan Africa that has achieved at least 30% growth in fertilizer use per cropped hectare over the past decade and which already started from a relatively high base . Several studies have been conducted in Kenya on adoption of improved maize seed and fertilizer on maize production. However, few microstudies on fertilizer adoption in general agriculture without focusing on a single crop have been conducted. In this study, therefore, we examine patterns in smallholder fertilizer use over time and estimate econometrically the determinants of fertilizer use in general agriculture. This is aimed at providing an empirical basis that would guide future fertilizer promotion policies and programs in Kenya and provide lessons for other SSA countries with regard to fertilizer use. By obtaining a clear understanding of farmer characteristics, institutional and geographic factors associated with fertilizer use in general agriculture, a platform is provided for policy makers to more accurately institute necessary programs that promote fertilizer use for agricultural productivity growth. The remainder of the paper is organized as follows: a brief historical perspective of fertilizer use in Kenya is presented in section 2. Section 3 presents a review of selected agricultural technology adoption studies. Data and methods are discussed in section 4. Section 5 presents and discusses findings while section 6 concludes with conclusions and policy implications.Fertilizer market in Kenya was liberalized in the early 1990s, leading to phasing out of all fertilizer subsidy programs.

Since then annual fertilizer consumption has progressively risen from a mean of 180,000 tons during the 1980s, to 250,000 tons during the early 1990s, to over 400,000 tons in the 2004/05 and 2005/06 seasons . In 2007, fertilizer consumption in Kenya stood at 451,219 metric tons. However, reports indicate that about300,000 tons of fertilizer was consumed in 2008 and this drop was due to civil disruption and the escalating prices of fertilizer in the world markets. Evidence suggests that growth in fertilizer consumption in Kenya is occurring on smallholder farms – it is not driven by large-scale or estate-sector agriculture. Growth in fertilizer consumption in Kenya is a phenomenon covering both food crops as well as cash crops such as tea, sugarcane, and coffee. However, as Ariga et al. note, the recent increases in world fertilizer prices combined with the civil disruptions experienced in early 2008 are likely to break the steady upward trend in fertilizer use that Kenya has experienced over the past 15 years .The growth in fertilizer consumption in Kenya has been achieved without subsidies. Three main arguments being advanced for the expanded use of fertilizer by small farmers in Kenya are: a relatively stable fertilizer marketing policy since 1990; increasingly dense network of fertilizer retailers operating in rural areas, leading to enhanced farmers’ access to fertilizer; and intense competition in fertilizer importing and wholesaling, creating pressure to cut costs and innovate in logistics. These arguments reveal the institutional factors that have contributed to expanded fertilizer use in Kenya. A micro-assessment of household and environmental factors associated with fertilizer use remains necessary to widen understanding of fertilizer adoption among especially smallholder farmers. This understanding can be a platform for designing informed strategies aimed at sustaining the momentum in fertilizer use in Kenya and promoting fertilizer use in other SSA countries.

Several adoption studies have been conducted in Kenya, most of them based on an initial desire to gather basic information about the use of modern crop varieties and inputs and to identify constraints to technology adoption and input use. A large number of these studies concentrate on cross-sectional analysis of the determinants of agricultural adoption at the farm level. The dynamics of the adoption process are not taken into consideration in cross-sectional analysis and the adoption process is represented as a snapshot in time. The coefficients may be biased since there may be a time-dependent element in the adoption decision. This section reviews some of these studies undertaken in the past. A study by Jayne et al., determined the national-level, region-specific, and household specific factors associated with smallholders’ use of improved maize technologies in Kenya and Zambia where over 25% of the farms use improved maize technology. Their study documented important factors that led to fertilizer and hybrid seed adoption on maize production. Among the factors identified included household characteristics such as education of head, distance to market, and regional differences. Though their study adopted a panel approach, the authors considered fertilizer adoption only on maize, excluding other crops such as coffee, tea and sugar cane whose production play a big role in household income in Kenya. Ariga et al.,large pot with drainage used household panel survey data to examine trends in fertilizer use on maize by smallholder maize growers in Kenya. The study employed Probit and Tobit models to identify factors that affect the decisions by maize farmers to participate in fertilizer markets and conditional on participation, their level of purchases. The study found that the dominant factor influencing smallholders’ decisions to use fertilizer on maize was location. The decision of households to purchase fertilizer for maize production was slightly related to farm size, and unrelated to household wealth. Proximity to fertilizer retailer was found to be an important influence on households’ decision to purchase fertilizer for maize production in the relatively low-potential areas. Proximity to fertilizer seller, however, had very little influence on the quantity of fertilizer purchased. This study considered only fertilizer use on maize and excluded crops such as tea, coffee and sugar cane which, in Kenya, are important drivers of growth in fertilizer use. In their study, Hugo De Groote et al. analyzed factors influencing adoption of maize technologies and fertilizer. Their study found that the proportion of farmers using improved varieties of maize had not changed but there was a positive tendency for the proportion of farmers using fertilizer on maize. They found that education, access to credit, access to extension and agro-ecological differences had significant influence on fertilizer adoption on maize. A study by Ouma et al. , using cross-sectional data, found agro-ecological differences, gender, manure use, hiring of labour and extension as statistically significant factors in explaining adoption of fertilizer and hybrid seed on maize production in Embu district.

The CIMMYT studies in Kenya and other East African countries examined adoption decision processes for maize seed and fertilizer technologies and showed that farmer characteristics such as age, gender and wealth are key to adoption decisions. In her study, Suri provided a succinct overview of the determinants of maize technology adoption in Kenya and showed that technology profitability, farmer learning as well as observed and unobserved differences among farmers and across farming systems were major determinants of adoption. Learning through social networks may also be an important determinant of technology adoption. Suri demonstrated that aggregate adoption rates may remain low or stagnant despite high average returns to new maize technologies, either because marginal returns to adoption are low, or because the farmers with comparative advantage in adoption have already done so. Mwabu et al. in their study on adoption of improved maize varieties and impact on poverty in Laikipia and Suba districts found that the price of maize, education level, and distance to roads are the main determinants of hybrid maize adoption by farmers. Karanja et al., applied a Tobit model on cross-sectional data to assess determinants of fertilizer adoption and use in Kenya. Their results indicated that fertilizer adoption and intensity of use was adversely affected by distance to fertilizer market and fertilizer price. Farmers closer to market tended to use more fertilizer. Farmers using hybrid maize seed used more fertilizer with the effect varying with agro-ecological zones.The study further noted that education, at post-secondary level, price of maize and extension positively influenced use of fertilizer on maize. Farmers with higher education tended to adopt and use more fertilizer on maize. This could be because they were able to use recommendations better or had a better ability to evaluate the difference fertilizer makes to productivity.The data for the study is obtained from a panel of households surveyed in 1996/97, 1999/00, 2003/04 and 2006/07 cropping years by Egerton University/Tegemeo Institute, with support from Michigan State University under Tegemeo Agricultural Monitoring and Policy Analysis Project . The sampling method for the panel of households was a mix of multistage and systematic. Twenty-four districts were purposively chosen to represent the broad range of agro-ecological zones and agricultural production systems in Kenya. Next, all nonurban divisions in the selected districts were assigned to one or more AEZs based on agronomic information from secondary data. Third, proportional to population across AEZs, divisions were selected from each AEZ. Fourth, within each division, villages and households in that order were randomly selected. A total of 1,578 households were selected in the 24 districts within seven agriculturally-oriented provinces of the country. The sample excluded large farms with over 50 acres and two pastoral areas. Households in Turkana and Garissa districts were not interviewed in the 2004 and 2007 surveys. This analysis is based on 1,275 households which formed a balanced panel for the four cropping years . The spread of the districts across the agro-ecological zones is presented in Annex 1. Ariga et al observes that a major advantage of panel data is that it overcomes problems of sample comparability over time. While in many countries there exist various farm surveys that can be used to measure patterns and trends in technology adoption over time, the comparability of these surveys is often compromised by differences in sampled households, locations, month/season of interview, recall period, and the way in which data is collected. The balanced panel on which the findings reported in this study are based provides a unique opportunity to track historical patterns in and explore determinants of fertilizer use for a consistently defined nationwide sample of small-scale farmers.This paper seeks to build on existing work on agricultural technology adoption in sub-Saharan Africa by assessing fertilizer adoption behaviour of farmers in Kenya over time. It is well understood that technology generation and development is an iterative process and the supply of technologies needs to be driven by demand from the users. Adoption studies are, therefore, important for the following reasons: to assess impacts of agricultural research; to aid in priority setting for research; and to provide information for policy reform to reduce constraints to adoption. This study reports historical patterns, based on the ten-year panel data, in smallholder fertilizer use to expose key trends. However, the historical patterns alone cannot provide information about the factors associated with fertilizer use.