An Economic Assessment of Banana Genetic Improvement and ...

SOCIAL CAPITAL AND SOIL FERTILITY MANAGEMENT IN UGANDA 97
Table 7.2 (continued)
Variable
Definition
Expected
effect
Mean
Standard
deviation
Social capital (Ω s )
Membership density Number of household members that belong to at least one association +/– 1.251 0.975
Leadership
heterogeneity
Continuous index measuring the degree of leaders heterogeneity in
terms of livelihood or education higher than most group members in
the village
+ 4.441 0.559
Norms of
decisionmaking
Net labor transfers
Net cash transfers
Net other item
transfers
Continuous index measuring the degree to which decisionmaking in
village associations is participatory
Value (thousand Ush) of labor obtained from the social network less the
value of labor supplied to the social network
Amount of cash (thousand Ush) obtained from its social network less
the amount of cash supplied to the social network
Value (thousand Ush) of other household items obtained from the social
network less the value of those items supplied to the social network
+ 6.114 0.4725
+/– 0.339 8.153
+/– –1.555 163.111
+/– 0.296 39.791
of the dependency ratio on use of mulching
and manure cannot be determined a priori.
The decision to invest in soil fertility
management practices is also conditioned
on farm characteristics (Ω F ), such as location,
physical factors of the land, and scale
of production. Elevation, a sample stratification
parameter, is expected to condition both
use of practices and perceptions. At higher
elevations where there are more slopes on
farms, the soil erosion potential is higher,
which could stimulate farmer’s perception
of the soil fertility problem and consequently
investment in soil conservation (Ervin and
Ervin 1982). Prior biophysical information
also indicates that banana productivity potential
is higher in high-altitude areas, which
provides further incentives to use soil fertility
management practices. Physical characteristics
of the land, such as the slope of the
farm and soil moisture retention capacity of
the banana plot, are hypothesized to directly
affect perceptions of the soil deterioration.
The slope of the farm represents the erosion
potential while the capacity of the soil to
retain moisture measures the capacity of the
soil to support the high demand of the banana
crop for water.
Therefore, low soil moisture retention
capacity of banana plots should influence
the demand for mulch and manure through
its influence on perceptions. On the other
hand, the counteracting effect of poor soil
moisture retention capacity on the productivity
of mulching or manure could lower
the incentive to incur costly investments.
Physical characteristics of land were measured
in qualitative form using a subjective
indicator reported by the farmers (Table
7.2). The drainage condition of the banana
plot is another land characteristic important
in banana production (Tushemereirwe et al.
2003). Bananas grown on poorly drained
soils are more vulnerable to leaf spot diseases,
which pose an exogenous risk in soil
fertility management decisions.
The scale of production (measured by the
banana mat count) reduces the fixed costs of
information acquisition per unit area, thereby
increasing the benefits from adoption (Feder
and O’Mara 1981). The scale of production
could also reduce the economic impact of
soil depletion on the household, as does the
landholding size. However, with respect to
mulching technology, landholding size increases
the access to mulch materials, while
the scale of banana production may act,
through fixed costs of information acquisition,
to influence use. These mechanisms
are conceptually different, justifying inclusion
of both farm characteristics in the
estimation.