How Do Quality Attributes of Banana Vary by Market Segments? A Hedonic Pricing Analysis

Regular intake of high-nutrient fruits is essential for healthy (Polsky and Garriguet, 2020; Liu 2003) and an active lifestyle (Sidhu and Zafar, 2018) as they are rich in vitamins and minerals (Better Health Channel, 2018; Slavin and Lloyd, 2012). It helps not only to prevent non-communicable ailments (Ruthsatz and Candeias, 2020) but also lowers the risk of chronic diseases such as stroke, cardiovascular disease, hypertension and Type 2 diabetes etc. (Liu et al., 2000; Bazzano 2006; Health Canada 2019; Li et al., 2014). The World Health Organization (WHO) estimated that cardiovascular disease alone will claim the lives of around 18 million people globally (Singla et al., 2022). Thus, the FAO and WHO recommend that adults consume at least 400g of fruits and vegetables each day, particularly if they are elderly (WHO, 2003).
       
Bangladesh has a per capita fruit consumption of only 35.78 g/day, while poor consumers intake just 15.19 g/day (BBS 2017), which is far below India’s 57.69 kg/person and worth noting as China and Canada which have the highest fruit consumption worldwide at 97.93 kg/person (FAO, 2021a). This clearly shows that Bangladesh consumes less fruit per person than other nations, particularly bananas, which are declining at a rate of 0.64% per capita while global consumption is rising at a rate of 0.22% (FAO, 2021 b). This is because the price difference between bananas and other fruits sold in the market is the main factor driving underprivileged consumers to purchase bananas.
       
Although Bangladeshis only consume 27.62 kg of fruit year on average, they rely heavily on bananas to prevent malnutrition, making them the second most nutritious food in the country in terms of kilocalories per decimal over the year (Fiedler, 2014). Bangladesh produces bananas only for domestic consumption (BBS, 2018), ranking 14^th globally (Hossain, 2014) among 112 countries where bananas are cultivated (UNICEF and WHO, 2017). Even in 2016-2017, the total banana production of Bangladesh was 0.8 million tons (BBS, 2018; FAO, 2021b), representing 20% of total fruit production and a market share of 36% (Fonsah et al., 2017).
       
The purpose of the study, therefore, is to identify the implicit price of banana quality attributes. The study will overcome the knowledge gap of the consumers to purchase bananas, which may overcome the health and environmental risks (Ikhajiagbe et al., 2021). Therefore, the study postulates three specific objectives namely; (a) To identify which quality attributes of banana, pay off the most, (b) To find out the buying behaviour of the different income-level consumer groups and (c) To calculate the marginal price of the quality attributes.
       
The study is important since bananas contain several bioactive compounds that may provide several health benefits (Singh et al., 2016; Sidhu and Zafar, 2018) and the consumption of bananas improves human health because bananas contain iron, fiber, vitamins and minerals (Choo and Aziz 2010; Kumar et al., 2012; Evans et al., 2020; Sora et al., 2023; Kukreja and Sharma, 2024). Bananas also contain several essential nutrients (Prodhan et al., 2017) and various antioxidants (Someya et al., 2002; Kanazawa and Sakakibara 2000) that protect against coronary heart disease (Hertog et al., 1993). The only solid food given to the infant is banana (Hossain 2014; Hossain et al., 2016). Nigerians eat bananas as a filling snack at home and while traveling (Emerole et al., 2013). The mixture of rice and ripe banana is one of the traditional and delicious foods in Bangladesh (Hossain 2014). Often referred to as the “fruit of the wise men” in India, bananas are currently regarded as one of the most significant fruit crops in the world (Saravanapandeeswari and Vanitha, 2018; Hazarika et al., 2021).
       
Literature on the hedonic pricing of bananas is not very strong (Basan 2017; Mugisha et al., 2008), though consumer preferences for banana purchase decisions prevail a lot. While Sporleder et al., 2014 demonstrated that trust plays a significant influence in customer choices; Basan (2017) observed that consumers, particularly high-income consumers in the Philippines, evaluate quality qualities and pay a price premium for bananas. The study added that low-and middle-income consumers accept bananas with or without blemishes. Banana size, taste, flavor (Ayinde et al., 2008), sweetness, color, flavor (Opara et al., 2007) and texture (Mugisha et al., 2008) which are also important for banana purchase and consumption decisions. The German consumers added that bananas with yellow-green colors are preferred over bananas with brown color (Symmank et al., 2018). On the other hand, consumers from Uganda concluded that they were willing to pay a 70-98% price premium for genetically modified bananas (Kikulwe et al., 2010). To the authors’ knowledge, the study of banana quality attributes in developing countries like Bangladesh does not exist. Therefore, the study confirmed that banana retailers may benefit if they consider the quality attributes of bananas in their selling strategies.
       
The next section explains the study's materials and methods, then moves on to results and discussion and finally conclusion.

The hedonic price model is a popular method of identifying the implicit price of a product’s quality attributes. The model was applied following the study of Waugh (1928) which first identified the price of vegetables (asparagus, tomatoes and cucumbers) based on some quality attributes, i.e., size, color and variety etc. The theory was later formally introduced by Court (1941) and Tinbergen (1951, 1956).
       
Lancaster (1966) used the model to make consumer choice decisions. He said “A product is a collection of quality attributes for which consumers buy the product and maximize its utility, subject to a budget constraint”. The following is the typical hedonic price model for calculating the implicit price of a product’s quality attributes (Lancaster 1966). 
 

 

 

       
Where,
p= Price of a product as the function of its quality attributes.
Zk= Vector of quality attributes.
x= Consumer utility is the function  of a numeraire good.
z= Purchase of the differentiated good.    
Where,
y= Total income of the consumer is assumed to choose a bundle of goods.
       
Based on the above stated standard hedonic price model, the following augmented model was used to calculate the implicit price of the banana quality attributes.
 

 

 
Here, log (p)= Logarithm of banana price.      
log(q)= Logarithm of quantity, a consumer purchase at a time.
a= Average price of banana if all things being equal.              
b, g, d and J= Indicate the estimated values of the parameters of quantity, variety (sagar or others), size (grade I or II) and the rest of the quality attributes such as color, wrinkle-free, damage-free, ripeness, traceability, chemical-free, hard flesh.
Uk = Unobserved banana quality attributes which are assumed as null for this study.
       
The regressors variables are measured by dummy variable except quantity purchased (in continuous form). The semi-log functional form of robust regression models were applied to find out the value of the estimated parameters.
       
At first, the parameters were estimated using the full sample. Later, the parameters were estimated for different consumer groups by income, banana variety and size. All the variables are interpreted following Halvorsen and Palmquist (1980). The validity of the model was tested by the Breusch-Pagan/ Cook-Weisberg (bp) test.
 
Field work
 
In total, 200 samples were collected from three different regions (Barishal, Khulna and Dhaka) of Bangladesh during the period of September-October 2020. The sampling areas are illustrated in the following Fig 1.


       
The samples were collected randomly using a semi-structured questionnaire. Before that, the questionnaire was pre-tested among 15 banana consumers and according to the pre-test, the questionnaire was finalized for this study. The questionnaire was divided into three sections: general inquiries about buying bananas, examination of banana quality qualities and customer demographics questions. The following Table 1 presents the demographic profile of the respondents.


       
Table 1 shows that 70% of the total respondents are from Barishal division and the remaining are from Khulna (14%) and Dhaka (16%). The adult age group has the greatest number of respondents. The reason is that the respondents are mostly from other occupations namely; students, drivers and housewives, etc. (52% of the total respondents). Around 18% of respondents are self-employed or business owners, with the remainder working in the private (23%) and government (7%) sectors. The table also shows that the maximum number of respondents’ income is low due to other occupation categories and represents at least 62% of the total respondents. The result is opposite to the study of Rasoli et al., (2021), where they found that there is a positive relationship between income and purchase of food products.
       
The table also shows that the average monthly expenditure on fruits is BDT. 2,760 and the average monthly expenditure on bananas is BDT. 304. The findings are consistent because people spent less money on fruits because the majority of their money was spent on necessities such as rice, fish, lentils and so on. The following Table 2 presents the summary statistics of the variables included in the hedonic price models.
       
Table 2 shows that consumers purchase at least seven bananas at a time. The price per stick banana is almost 6 BDT. The findings suggest that yellow bananas, which are ripe and have a sweet taste, are the main reason why consumers buy them. Consumers are less concerned about banana traceability because maximum people in Bangladesh are unaware of this attribute. Similar results are revealed in the estimation of the hedonic price models, where it is seen that traceability is not significantly important to the decision to purchase bananas for this study (see Table 3 and 4). However, consumers consider size and wrinkle-free in some context.

Eight regression models in total were estimated to determine the significance of quality attributes affecting consumer choices for bananas in Bangladesh. Table 3 below shows the estimated parameters of the choice to buy bananas for various consumer income categories, encompassing the entire sample size.
       
The first model shows the consumer purchase decision for the full sample size. The latter three models show the purchase decisions of low-, middle- and high-income consumers. The table shows that all the models are valid since the F-statistics are significant at 1-10% levels. The R-squared values illustrate how price variations are caused by several banana quality parameters. For example, the price of bananas would change by 42% if the variables included in the full sample size model changed by 100%. The result shows that quantity, variety, size, color, damage-free and hard flesh are significantly important for banana purchase decisions. P-values indicate that ripeness, wrinkle-freeness, traceability and chemical-freeness are not significant factors in any situation when it comes to banana buying decisions. However, the results may be different if a large number of sample sizes or more areas are considered in the scope of future research.
       
The results infer that consumers pay more for banana variety, size, color and damage-freeness while pay less for an extra stick and banana’s hard flesh for full sample size. The results are in line with the study of Hossain and Badiuzzaman (2021) especially in the case of Guava for size, color and Hog plum for quantity. The results are also consistent with the study of Hossain et al., (2022) with respect to the parameters explanation of color and extra stick of banana. Consumers pay at least 12% more if the banana is sagar, ceteris paribus. On the other hand, consumers pay least for an additional stick, discounted by at least 5% if other things remain constant. Consumers also pay less for hard flesh by at least 3%, ceteris paribus.
       
Similar results prevail in middle-income consumers, with the exception of the color attribute. High-income consumers consider only the variety of banana and are willing to pay 13% more for this attribute. Poor consumers purchased bananas more than the other two income groups’ consumers; at least 62% of the total respondents Table 1. Consumers are willing to pay more for size and variety and less for quantity. The similar results in the case of size and less for quantity are observed in the study of Hossain et al., 2021 for hog plum purchase decision in the southern part of Bangladesh. The following table 4 illustrates poor consumers’ buying decisions of bananas by their variety and grade.
       
The results show that all models are valid in terms of the p-values of F-statistics Table 4. With the exception of ripeness, traceability and chemical freedom, the models adequately predicted the price fluctuations of bananas. For example, at least 84% of the variation in banana price can be explained by changes in each of the independent variables included in the grade II model.
       
The results show that poor consumer buying decisions vary by banana variety and its grade. Consumers make different purchasing decisions for different banana varieties, such as sagar banana, which considers only a single attribute, namely, quantity, whereas other banana varieties, such as sabri, champa and so on, consider more attributes such as size and color. The price premium for the size of sabri, champa and so on is consistent with the study of Sarder et al., (2020).
       
The poor have different options when it comes to banana size. When making purchasing decisions for both grades of banana, consumers consider variety. The differences between grades I and II are observed in terms of wrinkle-free, damage-free and hard flesh. For example, consumers pay at least 27% more if the grade I banana is sagar, ceteris paribus. Consumers also prefer grade I bananas for their hard flesh quality and are willing to pay 3% more for it, whereas grade II bananas are preferred if they are wrinkle- and damage-free. Consumers pay 14% more if the grade II banana lacks wrinkle and pay 13% more for damage-free bananas, ceteris paribus. However, according to the respondents, the color of the banana-yellow in Bangladesh-generally reflected its ripeness. The results show that poor consumers purchase mostly other types of bananas due to their lower price. The results also show that consumers purchase grade I bananas due to their quality attributes rather than the grade II bananas.
               
The results aforementioned reveal that variety is the most important attribute for banana purchase decisions due to its availability all over the country. The result is consistent with the study of Emerole et al., (2013). Size and quantity are the second-most important attributes for the purchase decision of a banana, followed by damage-free, hard flesh, color and wrinkle-free. In Nigeria, size is one of the most important factors in banana purchasing decisions, while color is the least important (Ayinde et al., 2008). Poor consumers pay 27% more for sagar banana when its size is equal or more than 4 inches. Similar results were identified in the study of Basan (2017). In addition, the study revealed that consumers pay more for damage-free bananas. The study further revealed that color (yellow) positively influences the price of bananas (Mugisha et al., 2008) which is opposite to the study of Basan (2017). Basan also discovered that the price of bananas in the Philippines is adversely affected by green and green-yellow bananas.
 
Policy implications
 
The study suggests that the actors involved in the fruit supply chain can consider the factors influencing consumers purchase decisions. For example, the producer may consider the variety and the size of the banana to maximize their profits. The reason is that the majority of the banana growers have low-to-medium knowledge of banana cultivation technology (Thorat et al., 2014) and lack modern agricultural practices for producing quality bananas Fonsah et al., 2017. Therefore, the knowledge of farmers about the modern cultivation technique should be improved and it should be encouraged for them to adopt the new technology to produce more bananas (Hossain et al., 2015; Thorat et al., 2014, Basan, 2017) so that they can earn more profits from this fruit business.
       
The aratdars may consider the storage and transport facilities of bananas so that the fruit cannot be damaged. Therefore, the farmers are not willing to produce more bananas due to lack of proper storage facilities (Kamal et al., 2014). Similar findings were found by Ni et al., (2020), who also claimed that freshness has a major role in determining bananas market price. Retailers may take into account a number of attributes in order to divide the market into distinct customer groups and ascertain their needs. Finally, the government may monitor to discourage the use of any chemical for the protection of human health, sustainable environmental development through agricultural extension officers. The further research may be improved the quality of bananas with the coordination and integration between university, Bangladesh Agricultural Research Institute (BARI) or extension officials and research agencies.

Quality bananas are necessary since the poor consumers’ health may be worsening because they do not have access to alternatives in the Bangladeshi fruit market at the same price. Therefore, the aim of the study is to identify the implicit price of the banana attributes.
       
The study found that banana variety is the most important attribute for consumers purchase decisions. Size, variety, color, damage-freeness, wrinkle-freeness all increase the price of bananas, while hard flesh and quantity decrease the price. The findings show that poor consumers prefer other types of bananas due to their lower price and grade I for quality attributes. The study also revealed that poor consumers pay more for sagar variety and sizes of 4 inches or more, while paying less for an extra stick. The lack of damage and hard flesh of bananas distinguishes middle-income consumers from poor consumers. High-income consumers consider only the sagar variety for their purchase decision and pay 13% more. Consumer’s value other varieties quality attributes more than sagar’s. The results further revealed that consumers consider more attributes for a grade II banana than a grade I banana due to the latter’s lower quality.
       
The implication of the study is that actors involved in the banana supply chain may benefit from this study, considering the factors important for banana purchase decisions. The banana’s quality may be ensured if proper transport and storage facilities are improved. Therefore, the role of aratdars is highly important for selling bananas. The retailers also consider these factors to sell the bananas according to the consumers’ needs and preferences. The distinguished authority, such as BARI may supervise the production system to control the use of any residues for the production and distribution of bananas.

The contingent valuation method may be applied to rank the banana attributes. The study only shows the price variation of banana due to its quality attributes, lacks the consumption patterns of different regions. The study may be different if a larger sample size would be considered, a future scope of the research.

The authors acknowledged all the respondents who actively participated in the survey of this research work. The authors also thank the enumerators for their cooperation collecting the valuable information.
 
Disclaimers
 
The views and conclusions expressed in this article are solely those of the authors and do not necessarily represent the views of their affiliated institutions. The authors are responsible for the accuracy and completeness of the information provided, but do not accept any liability for any direct or indirect losses resulting from the use of this content.

The authors declare that there are no conflicts of interest regarding the publication of this article. No funding or sponsorship influenced the design of the study, data collection, analysis, decision to publish or preparation of the manuscript.