Department of Agricultural, Food, and Resource Economics Food Security Group

IDWP 125. Megan Sheahan, Roy Black, and T.S. Jayne. 2012. What is the Scope for Increased Fertilizer Use in Kenya?

EXECUTIVE SUMMARY:
Despite upward trends in fertilizer application rates on maize fields over the last twenty
years, there remains a perception in Kenya that fertilizer use is not expanding quickly enough
and that application rates are not high enough to reverse the country’s growing national food
deficit. In 2007, this manifested in the creation of a comprehensive multi-million dollar
fertilizer and improved seed subsidy and training program, the National Accelerated
Agricultural Inputs Access Program (NAAIAP), with the objective of raising food production
and farm productivity. However, little nationwide and longer term evidence exists to
determine whether higher fertilizer application rates are profitable for farmers and whether
they would have an incentive to continue using it on commercial terms after graduating from
the subsidy program.

Using five waves of nationally representative household survey data covering thirteen years,
this paper estimates the profitability of fertilizer application on maize fields and compares
with observed fertilizer use patterns over time. Marginal and average products of nitrogen are
estimated from a production function and disaggregated by district and soil type in order to
approximate local level agro-ecological conditions. In an environment where the real prices
of both fertilizer and maize have decreased since the late 1990s, relative input to output prices
have stayed fairly constant over the survey years, apart from a spike in fertilizer prices in
2007. Transportation costs of fertilizer, on the other hand, have decreased 35% over the
survey years given the proliferation of fertilizer retailers in rural areas, leading to a decrease
in the overall acquisition price of fertilizer.

By estimating economically optimal nitrogen application rates under both risk neutral and
risk averse scenarios and comparing with actual observed application rates, we find
households across Kenya have consistently and steadily adjusted their fertilizer use towards
optimal application rates over time. Over the entire sample, only about 16% of maize fields in
2010 were fertilized at levels less than 25% below our risk averse optimal values. This trend
is most pronounced in the Eastern and Western Lowlands areas where we find an appreciable
increase in the percentage of fertilized fields over time as well as an increase in the rate of
commercial nitrogen applied per hectare by fertilizer users, with a particularly large jump in
2010. Still, however, we estimate room for profitable expansion in these areas. In the High
Potential Maize and Western Transitional Zones, households see a gain in household income
from using fertilizer, however the last unit is generally at break-even profitable levels, or not
profitable at all, meaning households are applying fertilizer at optimal or slightly more than
optimal levels. Expanding fertilizer use in these areas is not a profitable strategy unless
coupled with complementary inputs and soil management practices.

Because relative prices do not show much variation over time, we also calculate two absolute
measures of fertilizer profitability, (1) the total revenue added from fertilizer application and
(2) the gain to fertilizer use at the margin. Results show that, despite an increase in fertilizer
application rates over time in some areas and the leveling off in others, the estimated total
revenue added from fertilizer application and the net gain to the last unit of fertilizer have
both eroded over time. With evidence from both relative and absolute profitability measures,
we find that tremendous additional expansion of fertilizer application rates on maize in
Kenya should not necessarily be sought after unless it is possible to raise the average physical
response rates of maize to fertilizer.