Abstract:

Banana is an important world food crop supporting the food security and livelihoods of
millions of smallholders in tropical countries of Africa, including Uganda. Despite the crop's
importance in Uganda, its productivity has been declining over time due to pests (banana
weevils and nematodes), diseases (black Sigatoka, banana bacterial wilt), soil fertility
decline, and socio-economic constraints (high costs of managing the crop, competition for
labour with other enterprises, marketing difficulties and low genetic diversity, among others).
The decline in banana yields has resulted in food shortages, thus putting consumers heavily
dependent on banana at risk of food insecurity, particularly in most rural areas of Uganda
where the crop is regarded a staple food.
In response to this trend, the National Banana Research Programme (NBRP) in Uganda
initiated a breeding programme in 1994 using a participatory plant breeding approach. The
programme has so far developed four new banana 'matooke' hybrid varieties (M2, M9, M14,
and Ml 7). These varieties have been under evaluation in different agro-ecological regions of
Uganda since 2008 with Mbwazirume (a traditional variety) as a local check. Despite the
research and extension efforts to popularise these hybrids in Uganda, to date, no attempt had
been made to document consumer acceptance and the likelihood of farmer adoption of these
hybrid bananas. Little is known about the socioeconomic factors that influence farmers’
early-stage adoption of the banana hybrids, farmers’ preferences of the varieties, consumers’
willingness to purchase the bananas of the hybrids when found on the market, and the
likelihood of these hybrid varieties contributing to solving rural household food insecurity
and low incomes.
Given this context, the objectives of this study were to: (i) analyse farmers’ preferences
regarding varietal attributes and assess how these preferences, along with farm and farmer
specific characteristics, determine the variations in the levels of early stage adoption of
hybrid banana varieties in Uganda with the application of factor analysis and a Zero-Inflated
Poisson (ZIP) regression model; (ii) analyse the effects of farmer characteristics, variety
attributes and agro-ecological conditions on banana variety choice in Uganda, with the
application of a multinomial logit model; and (iii) determine the consumption characteristics
and sensory attributes that are most important in predicting the likelihood of consumers’
purchase of hybrid banana varieties using binary logit models. The research focused on four
regions of Uganda (namely, Mid-Western, Central, Western and Eastern) representing six
major agro-ecological zones, including the Lake Albert crescent area, Lake Victoria crescent,
Western highlands, Southern highlands, South-east and Eastern agro-ecologies where the
National Banana Research Programme of the National Agricultural Research Organisation is
evaluating the new hybrid banana varieties. The data were collected from 454 farmers that
included participating (149) and non-participating (305) respondents, and 908 consumers that
participated in farmer field days to evaluate various hybrid banana varieties from across the
four regions of Uganda.
The results show that, compared to Mbwazirume, four of the hybrids (M2,M9,M14 and Ml 7)
are preferred in terms of production characteristics (resistance to Sigatoka, weevils,
nematodes, tolerance to poor soils, good bunch size, and sucker production) but are regarded
as inferior in terms of consumption characteristics (taste, colour when cooked, and flavour).
Field observations suggest that farmers’ preferences for production attributes dominate in
their variety choice decisions. The hybrid M9 is regarded as having a relatively good
performance with respect to most of the production and consumption characteristics.
The results suggest that land constraint, taste and regional location (central region compared
with eastern region) were negatively associated with hybrid variety choice while perceptions
that hybrid bananas could reduce food insecurity and enhance tolerance to pests and diseases
were positively associated with probabilities of hybrid variety choice. Probabilities of
choosing hybrids for food security increase in favour of M2 (by 0.06) and M9 (by 0.28), and
decrease for M2 (by 0.23), M9 (by 0.07) and M14 (by 0.09) due to unfavourable taste relative
to Mbwazirume. Among the consumers, the study suggests that hybrids M2 and M9 were not
significantly (P>0.05) different in terms of taste, flavour, texture and colour, while
differences were observed between M14 and M17 when introduced on the market. Hybrid
M14 is regarded as better than M17 in terms of taste, flavour and texture.
The results show that the Eastern region would pay significantly higher prices for the hybrid
varieties compared to other regions of Uganda. This result could be attributed to the limited
availability of cooking banana varieties in this region. The results suggest that age, education,
good taste, flavour and texture were the most important factors that are likely to positively
influence the purchase of most of the hybrid banana varieties.
The results further suggest that 41% of the host farmers were having more than sufficient
food supplies, while 85% mentioned that their household food situation had substantially
improved for the better over the past 3-5 years. The results show that 78% of the host farmers
perceived their quantity of food supplies at household level had increased compared to a
period before having access to hybrid banana varieties. This clearly shows that adoption of
these hybrid banana varieties would enhance rural food security in the country.
The results show that the majority of the respondents were planning to expand production
with hybrids M2, M9. M14 and Ml7. High yield, food security and income generation,
resistance to diseases and pests, drought tolerance and the longevity of the hybrid plantations
were the reasons given by farmers for their preference for the new hybrid varieties. The study
has demonstrated that farmers have relevant knowledge that can be used in setting banana
breeding priorities so that they can select varieties that potentially have traits of their
preferences. Therefore, involving farmers in the early stage evaluation of new technologies is
likely to shorten the time required for evaluation and adoption of the preferred varieties.
Future breeding efforts should target attributes like bunch size, good taste, soft food and
agronomic characteristics (including early maturity, performance in a good season, high
yielding, plant height and sucker production). Efforts should be made to promote and make
available planting materials to more farmers particularly of the most preferred hybrids like
M9. The Eastern region of Uganda could be targeted, especially with hybrids M9 and M14
given the scarcity of planting materials in this region and farmers’ willingness to offer higher
prices for the hybrid bananas.
Further work needs to be done to determine the dynamics of adoption and dis-adoption of
these hybrids, and the impact of hybrid bananas on household income and food security. A
panel data study could also be conducted to extend the dimensions of the current study and
allow for the control of unobserved effects that remain relatively fixed over time at the
household and community levels.