An Economic Assessment of Banana Genetic Improvement and ...

4 CHAPTER 1
practices in Africa are inconclusive in two
respects. Not only is their ability to encapsulate
what happened in the relevant time
period questionable, but their relevance to
emerging gene-based technologies in today’s
societies may also be limited.
Past assessments have documented numerous
examples of successful innovations
in African agriculture (Dommen 1988; Gilbert
et al. 1993; Sanders, Shapiro, and Ramaswamy
1996; Byerlee and Eicher 1997;
Haggblade 2004), but often their impacts
have been short-lived or difficult to substantiate
in quantitative terms at a national or
regional level of analysis. Once a technology
has diffused among farmers and farming
systems, disengaging the effects of
genetic change from related inputs and environmental
factors is a well-known challenge
(Alston, Norton, and Pardey 1995; Morris
and Heisey 2003). For improvements in pest
and disease resistance, establishing the scenario
that represents the absence of genetic
change is especially challenging when it
requires the estimation of farmers’ yields
over wide expanses in the presence of losses
that are not observable. Experimental data
on crop-yield losses reveals the severity of
the disease or pest when it occurs but not its
incidence across geographical regions and
farm-management regimes. Most often, despite
questionable assumptions, expert opinions
are combined with experimental data
in a sensitivity analysis that generates a
range of estimates of the percentage of crop
yields that would have been lost had the
improved seed technology not been adopted
(Marasas, Smale, and Singh 2003).
Broader social, political, and economic
contexts mediate the impacts of seed technological
change—which evolve over time
but may also change abruptly. For example,
the state-based public research and delivery
systems that supported the diffusion of
modern technologies during the decades
following independence of many African
nations subsequently proved to be fiscally
unsustainable. Once dismantled, some statebased
systems have been partially replaced
by a patchy mosaic of firms and nongovernmental
organizations. In other cases, former
“control” policies have continued in a more
contracted way, excluding smallholder farmers,
while continuing to benefit politically
important farming interests (Jayne et al.
2002).
Over the same period, the relative importance
of agricultural income in the income
of poor households has declined, so
that the effects of seed technological change
on employment and poverty may not be of
the magnitude they once were (Meinzen-
Dick et al. 2007). In Africa, market liberalization
has progressed unevenly; farm families
often cope with high and variable input
as well as output prices, striving to meet
their cash needs through numerous sources
in an increasingly monetized economy
(Bryceson 2002).
The failure of market liberalization has
been attributed to the failure of policies to
stimulate price signals to producers or to
farm-level constraints that effectively block
a supply response (Carter 2000). Encouraged
by donors, governments have sometimes
pursued a crisis-to-crisis approach to
agricultural policy in food crops, leading to
vacillating input–output price ratios for
smallholders and inconsistent technology
recommendations (for example, Heisey and
Smale 1995, revisited a decade later by
Smale and Jayne 2003). Policies that contributed
to apparent success in one period
later led to decline in smallholder use of the
technology.
In the meantime, the number of institutional
actors has expanded, and the configuration
of institutions involved in agricultural
R&D, as well as technology provision to
farmers, has changed. The productivity
gains of the Green Revolution occurred in a
narrow range of crops and technologies, financed
by commodity-oriented, supplydriven,
public national research programs
backed by international agricultural centers
(Byerlee and Echeverría 2002). “Orphan”
crops, such as the cooking banana, have
fallen largely outside the mandates of the