Advances in Crop Science and Technology
Open Access

Commercialization; Double hurdle; Households; Market-participation; Onion

Introduction

Ethiopia's economy depends primarily on agriculture, with smallholder farmers cultivating 95% of the country's farmland and producing 90% of its total agricultural production (Kusse et al., 2022). Agriculture is an essential driver of economic growth in Ethiopia. Crop and livestock production account for roughly 65% and 25% of agricultural GDP (ITA, 2024). Agriculture contributed 20.6% to poverty reduction, 37.2% to GDP, a78% to export income, and 75% to employment opportunities (WB, 2017). The GDP contribution of agriculture is growing in absolute terms over time, although the sector's share of the national GDP has been declining over time (Trending Economics, 2024). Ethiopian agriculture is dominated by small-scale farming. Around 67.5% of farmers have farmland of less than 4 hectares, and the average small farm size is 0.9 ha. Only 1.8% of farming households are operating on a farm size of more than 15 hectares (Kiru, 2019). Ethiopia's overall economy is dependent on the growth of the agricultural sector, to which smallholder farmers make a significant contribution to the country's overall agricultural growth. The movement of the agriculture sector depends entirely on what is happening in the smallholder subsector (MoARD, 2010; NBE, 2019) [1,2].

Agricultural commercialization refers to the increase in the proportion of the supply of agricultural output that is sold, instead of the quantity of agricultural output used by farm households for home consumption (Olwade et al., 2015; Minot et al., 2022). There are numerous perspectives and definitions of agricultural commercialization. It can be viewed as either a static or dynamic process over time [3]. The static form of smallholder farm commercialization could be seen as a measure of the strength of the linkage between farm households and markets at a given point in time. This household-to-market linkage could relate to output or input markets, either in selling, buying, or both, including labor (Moti et al., 2009; Tamirat et al., 2023). Considering farm commercialization as a dynamic process, it could be seen as a process in which the speed of the share of outputs sold and inputs purchased changes over time at the household level. According to this principle, agricultural commercialization occurs when enterprises involved in agriculture and/or the agricultural sector as a whole increasingly rely on the market for the sale of produce and for the acquisition of production inputs (Poulton, 2017). The process of agricultural commercialization occurs when agricultural enterprises and the agricultural sector as a whole rely more on the market to sell their produce and buy labor and other production inputs (Poulton, 2017) [4]. Commercialization of smallholder agriculture refers to the transition from subsistence to market-oriented farming, which can result in increases in production, income, and employment, as well as a decrease in poverty. Agricultural commercialization also improves the food supply in urban areas, with broader growth and welfare effects (Barrett, 2008; Carletto et al., 2017; Ogutu and Qaim, 2019).

Vegetable production, which involves commercial state farms, private commercial farms, and smallholder farming, is a vital economic activity in Ethiopia (Hagos et al., 2018). Onion (Allium cepa) is one of the most significant vegetable crops worldwide. Onion has mainly grown as a food source, is used as a cousin, and is a valuable addition to various dishes. In Ethiopia, it is grown mainly for its bulb, which is widely used for its flavoring properties, daily stews, and other applications in vegetable food preparation (Goldman, 2011; AgroBIG, 2016) [5]. Because of these significant advantages, onion production is increasing in the country's many agro-ecologies in small-scale production systems, being one component of commercialization and a daily source of income for both rural and urban populations (Muluneh et al., 2019). Onion is an important economic center in Ethiopia because of its ease of cultivation, higher yield per hectare, and the irrigation system that increases onion production from time to time. In Ethiopia, the total area under onion production was about 38,952.58 ha, of which 3,460,480.88 tons were produced in 2020/2021, with an average yield of about 8.8 t ha-1. However, productivity is significantly lower than the global average of onion productivity, which is 18.8 t ha-1 (CSA, 2021) [6].

Wolayita and Gomo zones are among the potential onion production zones of South Region Ethiopia. Despite the livelihood contribution of onion commercialization, so far as the author’s knowledge is concerned, in the study areas, there have been no similar studies on the commercialization and level of commercialization of onion producers in study areas. To fill this gap, this study aimed to investigate the determinants of onion commercialization and the level of commercialization [7]. In addition, some related studies conducted in different areas did not conduct model specification tests to select an appropriate model for their dataset. For example, Taye et al. (2018) used the Heckman two-stage model to analyze the determinants of commercialization of smallholder onion farmers in Fogera district, Ethiopia; Engida et al. (2021) used the Tobit model for analyzing commercialization and intensity of commercialization of sorghum in Southwest Ethiopia; Dubale et al. (2021) used a double hurdle model to analyze the commercialization level and determinants of market participation of smallholder wheat farmers in northern Ethiopia; Ater et al. (2021) analyzed the Factors Influencing the commercialization of Horticultural Crops Among Smallholder Farmers in Juba, South Sudan Using Tobit Model; Zelalem et al. (2023) used the Double Hurdle model to analyze the commercialization and level of commercialization of teff growers and determinants in west Ethiopia [8]. However, none of these studies conducted model specification tests to select a model against Tobit or Heckman's two-step double hurdle model. Moreover, Abdu et al. (2016), used a multiple linear regression model to analyze the determinates Smallholder Commercialization of smallholder farmers for the Coffee-Spice Based Farming System of South West Ethiopia, although the data set has zeros or missing values for its dependent variable household commercialization index. For a dependent variable with several zeros and missing observations, multiple linear regressions produce inconsistent and biased parameter estimates (Woodridge 2016). Thus, to fill these research gaps, this study used appropriate model specification tests of limited dependent variable models to investigate the determinants of onion commercialization and the level of commercialization of onion production in the study area [9,10].

Methodology

Description of the study area

The study was conducted in major onion-producing regions of the South Region of Ethiopia. The study was conducted in the largest onion producers in the Dugna Fango district of Wolayita zone and Mirab Abaya district of Gamo zone in the South regional state of Ethiopia. Mirab Abaya Wereda(district) is located 225 km south of Hawassa, covering a total area of 1,405 km². Of this area, 17,437 hectares are used for farming. The wereda's elevation ranges from 1,100 to 2,900 meters above sea level [11]. It receives an average annual rainfall between 800 and 1,600 mm, with temperatures typically ranging from 24°C to 30°C (Direslgne et al., 2016). Duguna Fango Woreda is located 42 km east of the zonal city Wolaita Sodo, 73 km southwest of Hawassa, and 300 km south of Addis Ababa. Geographically, the woreda lies between 6°45’00’’ and 7°5’00’’ N latitude, and 37°58’00’’ and 38°8’00’’ E longitude (Aklilu et al., 2020) [12].

Sources and method of data collection

The survey was conducted in Gomo and Wolayita zones, South Regional State of Ethiopia. For the study, both primary and secondary data were collected [13]. The field survey was conducted in 2019 to collect data from primary and secondary sources. Primary data were collected from randomly selected onion producers producing farm households by using a structured questionnaire and employing trained enumerators under the supervision of the researcher. The questionnaire was pretested and amended based on the feedback obtained to ensure validity and reliability. Before the commencement of the survey, training was provided to enumerators. Then data were collected using these trained enumerators from March to June 2019. Secondary data were collected from the Central Statistical Agency (CSA), Office of Agriculture and Rural Development, FAO, International Research Institution Report, and Online publications [14].

Sampling procedure and sample size

A multi-stage random sampling technique was employed to select the study locations and onion-producing households. In the first stage, the Wolayita and Gamo zones of the South Region were selected purposively based on their potential for onion production. The Second Mirab Abaya district of Gomo zone and Dugna Fango districts of Wolayita zone were selected based on their highest potential for onion production. Thirdly two onion-producing kebeles from each district, that is Bilate Chericho and Bilate Eta kebeles of Duguna Fango district and Kola Muleta and Yayke of kebeles of Abaya district were randomly selected. Finally, 160 onion producer-sampled households were selected using a simple random sampling method using probability proportion to size [15].

Method of data analysis

The collected data were analyzed using descriptive and econometric methods of data analysis. For the descriptive analysis, mean, frequency, and percentages were used. Inferential statistical tests such as t-test and chi-square test were used for the existence of any statistically verifiable differences among households participating in the onion market and their counterpart non-market participants. Under econometric analysis, double-hurdle model was used to estimate market participation decisions and the level of commercialization of onion producers [16].

Model specification

Following the work of Von-Braun (1994), the Household Commercialization Index (HCI) formula is given as:

A commercial index value of zero implies non-commercialization of onion while the closer it is to one, the higher the degree of onion commercialization.

Limited dependent variable models (Heckman two-step models, Tobit, Double hurdle) were used in the study of crop market participation. Heckman’s two-step model estimation of the data has the mills’ lambda, which is insignificant, which indicates there is no selectivity bias, which implies that Heckman is not an appropriate econometric model as compared to other limited dependent models [17]. Hence, the Tobit model and double hurdle models were compared using the model-specified test (Komarek, 2010). Based on the model specification test result as indicated results and the discussion section of limited dependent variables model specification (Heckman two-step models, Tobit, Double hurdle), the double hurdle model is appropriate for this dataset as compared to the Tobit regression model [18].

According to Wooldridge (2002), the Tobit model assumes that the household’s decision to sell and how much to sell if the sale occurs is determined by the same mechanism. With the model specification test discussed later below and shown as well in the model result, Tobit does not fit this research dataset following the specified test. Craggs hurdle fittest model for this dataset. Cragg’s model has two hurdles to overcome before observing the positive value of the dependent variable [19]. First is the decision to be made about whether to consider a change or not. Then, a decision on the amount of the change is taken (Cragg, 1971). According to Newman et al. (2001), the first hurdle involves the choice of to sell or not to sell onion production (market participation decision), whereas the second hurdle concerns the level of commercialization of the producer choices (quantity of sales decision). It indicates that a producer makes two decisions concerning the sale of onion production [20].

According to Humphreys (2013), the standard likelihood ratio test can be used to test the double-hurdle model against the Tobit model since the Tobit model is nested in the double-hurdle model. That is, the Tobit model can be derived from the double-hurdle model by restricting the parameters of the probit model to be equal to the parameters of the truncated regression [21].

If LLprobit is the log-likelihood of probit model, LLtruncreg is the log-likelihood of truncated regressions and LLtobit is the log-likelihood value of Tobit model. Then, the likelihood ratio test (LR) can be carried out as follows:

The test statistic has a χ2 distribution with degrees of freedom equal to the number of parameters that are included in the regression (Tobit = truncated = probit), plus the intercept. With the null hypothesis, the Tobit model is a better fit than the double-hurdle model. However, on the other hand, rejection of the null hypothesis means that the double-hurdle model is a better alternative to the data [22].

According to Moffatt (2005), the equations for the double huddle can be written as:

First hurdle that presents the onion market participation decision is expressed as (Equation 2):

The second hurdle, which represents intensity commercialization, is modeled as a truncated regression as follows (Equation 3):

= if > 0, or di = 1; = 0 if ≤ 0,

Where; i represent the i^th household head; and are vectors of explanatory variables; is the latent or unobserved market participation decision; and are the corresponding vectors of parameters to be estimated; is the observed amount of commercialization in the market; and is the latent or unobserved amount of commercialization to the market; and and are uncorrelated normally distributed error terms for both decisions, respectively [23].

According to Humphreys (2013), the Cragg hurdle assumes no correlation between and () is estimated by the following likelihood function (Equation 4):

The likelihood of the probit model and the truncated regression model under the assumption of independent error terms is the likelihood of the Cragg hurdle model. In this case, the first two terms on the left-hand side are the probit model for participation and the third term is a truncated regression model [24].

Definition of variables and hypothesis

Dependent variables

Market participation (MP)

It is a dummy variable that indicates the participation of the household in the market that is regressing in the first step of the two-step estimation procedure. If the household participates in the onion market, it takes a value of one. However, it takes a value of zero for households that do not participate in the onion market during the production season [25].

Commercialization index of households (HCI)

It is a continuous variable in the second step of the selected model. It is measured in the commercialization index and represents the actual level of commercialization of onions marketed by farm households during the production season.

Explanatory variables

Explanatory factors include demographic, socioeconomic, and institutional factors that affect onion commercialization and the level of commercialization. Table 1 presents a list of explanatory variables expected to affect onion market participation, the level of onion commercialization, and the hypothesized direction of association with the dependent variables. They were hypothesized based on the reviewed literature and economic theory (Table 1 & Appendix 1).

Table 1:Variables and Hypothesis.

Appendix Table 1: Descriptive statistics of continuous variables.

Results and Discussions

Descriptive statistics

This section provides an analysis of the demographic, socioeconomic, and institutional factors affecting onion producer households. The pertinent details are presented [26].

Age: The minimum and maximum age of household were 20 years and 64 years, respectively, while in terms of onion market participation, the average age of household head was 39.26 years for market participants and 45.5 years for non-market participants. The t-test result shows that the mean difference was statistically significant at 1% level, indicating that there is a variation in households in terms of the average age of household heads between onion market participants and non-participants (Table 2 & Appendix 2).

Table 2: Summary of continuous variables.

Appendix Table 2: Heckman selection model two-step estimation results.

Education level: The minimum and maximum education levels of household heads were 0 and 12 years of schooling, respectively, while in terms of onion market participation, the average education level of households was 6.96 years of schooling for market participants and 3.83 for non-market participants. The t-test result shows that the mean difference was statistically significant at 1% indicating that there is variation in the household head in terms of average education level of household heads between market participants and non-participants [27].

Family size: The minimum and maximum family sizes of households were 1 and 10, respectively, while in terms of onion market participation, the average household size was five for market participants and seven for non-market participants. The t-test result shows that the mean difference was statistically significant at 1% level, indicating that there is variation in households in terms of average family size between market participants and non-participants.

Farming experience: The minimum and maximum farming experience of households was two years and 40 years, respectively, while in terms of onion market participation, the average household farming experience was 17.24 for market participants and 17.42 for non-market participants. The t-test result shows that the mean difference was insignificant, indicating that there was no variation in households in terms of average farming between market participants and non-participants [28].

Frequency of extension contact: The minimum and maximum frequency of extension contact of households was 0 and 5 times, respectively, while in terms of onion market participation, the average frequency of extension contact was 2.51 for market participants and 1.21 for non-market participants. The t-test results show that the mean difference was statistically significant at 1% indicating that there is a variation of households in terms of the average frequency of extension contact between market participants and non-participants [29].

Livestock Ownership: The minimum and maximum livestock ownership in terms of tropical livestock units for households were 0.195 and 17.21, respectively, while in terms of onion market participation, the average livestock ownership was 9.18 for market participants and 8.61 for non-market participants. The t-test result shows that the mean difference was statistically insignificant, indicating that there was no variation in households in terms of average livestock ownership between market participants and non-participants [30].

Land allotted to onion: the minimum and maximum land size allotted for onion production by households was 0.125 hectares and 1 hectare, respectively, while in terms of onion market participation, the average land allotted to onion production was 0.57 hectare for market participants and 0.49 hectare for non-market participants. The t-test result shows that the mean difference was statistically insignificant, indicating that there was no variation in households in terms of average land size allotted for onion production between market participants and non-participants [31].

Distance to the nearest market: the minimum and maximum distance to the nearest marketplace of households was 0 kilometers and 3 kilometers, respectively, while in terms of onion market participation, the average distance to the nearest marketplace was 1.14 kilometers for market participants and 1.77 kilometers for non-market participants. The t-test results show that the mean difference was statistically significant at 1% level, indicating that there was variation in households in terms of the average distance to the nearest market between market participants and non-participants (Table 3 and Appendix 3).

Table 3:Summary of dummy variable.

Appendix Table 3: Double hurdle model result.

Access to training: The results in Table 3 show that 74.38 % of households had access to training, whereas 25.63% of households had no access to training on improved onion production. As shown in Table 75% of market participants had access to training, whereas 25% of market participants had no access to training. On the other hand, out of non-market participants, 70.83% of non-market participants had access to training while 29.17% of non-market participants had no access to training. The results of the chi-square test show that access to training was statistically insignificant, indicating that there was no association between market participation and access to training for households [32].

Econometric model

As mentioned in the research method of model specification, double hurdle models were selected based on the model specification test. Accordingly, the Heckman selection model was not appropriate for this dataset since IMR is insignificant (has a p-value of 0.123), as indicated in the appendix. Then, the double-hurdle model is evaluated against the Tobit model specification [33]. The test statistic for the log-likelihood ratio test of onion commercialization is (LR = 250.93) which by far exceeds the critical χ2 value of 23.209 at the 1% level of statistical significance and 10 degrees of freedom. This reveals that the double-hurdle model is appropriate against Tobit. A full double hurdle model is appropriate for the dataset if there is a correlation between error terms of participation and level of participation in the onion market (Humphreys, 2013; Engel and Moffatt, 2014). For this data full double hurdle is not appropriate since the IMR is insignificant (has a p-value of 0.13), indicating that there is no correlation between the error terms (see Appendix). Thus, the Cragg hurdle model of the double-hurdle model was appropriate for this dataset [34].

Determinants of onion market participation

The results for the determinants of market participation are estimated using the probit model, and the first step of the double hurdle is displayed. The likelihood ratio chi-square (LR chi-square) value of the probit model is 54.71 is statistically significant at 1% indicating that the explanatory variables in the model explain the probability of participating in onion markets. Out of the nine explanatory variables included in the model, four variables that were found to significantly influence the probability of participation in the onion market of producers in the study area are discussed as follows.

Family size

Family size has negatively affected the likelihood of onion market participants at the 1% level of significance. The marginal effect shows that an additional member increase in family size in the family decreased the probability of onion producers’ market participation by 4.1%. This indicates that as the number of family members increases, farm households are more concentrated on consumption rather than production for the market due to a lack of sufficient onion surplus production for the market. This is due to the large number of children who contribute more to consumption rather than labor for the production of onions. This result is in line with the findings of Arega et al. (2008) which stated that a larger household is likely to consume more output, leaving smaller and decreasing proportions for sale; Gebreslassie et al. (2015), family size had negative and significant association with the market participation of the smallholder wheat farmers; Guta et al. (2020) family size negatively affected vegetable market participation; Nigus and Tsegaye (2022), found that family size negatively and significantly affected Avocado market participation. These findings are in contrast to the findings of Osmani and Hossain (2015), who found that family size has a positive effect on smallholder market participation, and Banchamlak and Akalu et al. (2022) found that family size is positively associated with farmers’ likelihood of participating in vegetable market supply in the Yayo and Hurumu districts of Ethiopia [35].

Farming experience (years)

Farming experience positively affected the probability of onion market participants at the 5% level of significance. The marginal effect shows that an additional year increase in the farming experience of the household head increases the probability of onion producers’ market participation by 0.7%. This is because farmers with more farming experience have greater awareness and knowledge regarding the production and marketing of onions, as compared to their counterpart onion non-producer farmers; therefore, they are more likely to take the risk of production and marketing and participate in the marketing of onions. This result is inconsistent with Tesfaye (2021), who found that farming experience positively affected onion commercialization in Ethiopia [36].

Frequency of extension contact

The frequency of extension contact positively affected the probability of onion market participants at the 1% level of significance. As shown in Table 3, an extra day of extended visits by extension workers increases the probability of onion producer participation in the market by 3.6%. This is because extension agents consult farmers on modern onion production methods and provide information on market availability and new and better varieties that enhance productivity. This result agrees with the studies conducted by Tesfaye (2021); Banchamlak and Akalu (2022) found that extension contacts positively affected onion and vegetable market participation, respectively [37].

Distance to nearest market (km)

Distance to the nearest market negatively affected the probability of onion market participants at the 5% level of significance. The marginal effect shows that an additional kilometer increase in distance for the household decreases the probability of onion producers’ market participation by 4.2%. This is because the further is a household from the onion market, the more difficult and costly it is to get involved in the onion market. The closer a farmer is to the market; the easier it is to take the products to the market because the farmer may not incur a high transportation cost. This result is in agreement with the study of Guta et al. (2020), who reported that distance to the marketplace has a negative and statistically significant influence on the commercialization of vegetables. Similarly, Berhanu et al. (2009) found that a large increment in the distance of the nearest market causes a reduction in the probability of participating in the maize market. Gizachew (2005); Holloway and Ehui (2002) also found that negative relationship between distance to the market and the probability of participation in the milk market [38].

Factors affecting the level of onion commercialization

To determine the factors influencing the level of onion market commercialization, a truncated model was estimated in the second step of the double-hurdle model equation. The Wald chi-square value of the truncated regression model is 67.44 is statistically significant at 1% indicating that explanatory variable(s) in the model explain the level of commercialization of onion. Education level, family size, frequency of extension contact, land size allotted for onion production, and distance to the nearest market were found to have significantly influenced the level of onion commercialization in the onion market (Table 4).

Table 4: Regression result for double hurdle model of onion commercialization.

Education level

The education level of household heads affected onion producers’ level of commercialization positively and significantly at the 1% significance level. The result of the marginal effect revealed that a one-year increase in the level of education of the household head increased the level of commercialization in the onion market by 1.2%. This is because, as the level of education of household heads increases, the capacity to analyze and plan profitable types of farming business increases. The results of this study are consistent with those of Tadele et al. (2017), Addisu (2018), Agerie et al. (2020), and Asfaw et al. (2024), who found that the level of formal education increased the level of commercialization of wheat, teff, maize, and vegetables.

Family size

Family size is negatively and significantly affected the level of commercialization at a 10% level of significance. The marginal effect shows that an increase in the family size decreases the level of onion commercialization by 1.3%. This is because a large family needs more onion to consume and less amount remains for sale as compared to small family-size households. This finding is consistent with that of Gani and Adeoti (2011): Musah et al. (2014), and Zelalem et al. (2023), who Confirmed that the household‘s level of market participation decreased because of increased family size. These are in contrast to the findings of Banchamlak and Akalu (2022), who reported that family size positively and significantly affected the vegetable market supply [39].

Frequency extension contact

The frequency of extension contact positively and significantly affects the level of onion commercialization at the 1% level of significance. As shown in Table 3, an extra day of extended visits by extension workers increased the level of onion producer participation in commercialization by 3%. The positive and significant relationship of frequency of extension contact indicates that extension contact improved onion farming households' ability to acquire new technologies and capacity of production, which in turn improved productivity and thereby increased commercialization of onion. This result is in line with the study by Gezahegn et al. (2021), who found that the frequency of extension contact positively and significantly affected the quantity of onion and tomato sold to the market. Similarly, Kassa et al. (2020), Teklebrhan et al. (2020), and Nigus and Tsegaye (2022) find that extension contact positively and significantly affects banana, onion, and avocado levels of market participation.

Land size allotted for onion production

Land size allotted for onion production positively and significantly affected the level of onion commercialization at the 5% level of significance. One-hectare increase in land allotted to onion production increased the level of onion producers’ level of market commercialization by 9.9%. Households who allocate large plots of land for onion cultivation can produce more onions in surplus production beyond household consumption, thereby enabling households to supply more onion products to the market. This result is in line with Taye et al. (2018), who found that land allotted to onion production positively affects the amount of onion supplied to the market. Similarly, Ibrahim et al. (2021) found that the land allocated for tomato production positively affects the level of market participation.

Distance to nearest market (km)

Distance to the nearest market negatively and significantly affects the level of onion market commercialization at the 10% level of significance. The marginal effect shows that an additional kilometer increase in distance for the household decreases the level of onion producers’ market commercialization by 2.3%. This implies that the farther from the marketing center, the lesser the household’s tendency to participate in the market. The more likely reason may be the high transportation cost of products per unit volume of the products. As a result, households may be discouraged from participating in the market because of the high transportation cost of transporting farm produce to the marketplace. This result is consistent with the studies by Efa et al. (2016), Kassa et al. (2020), Ibrahim et al. (2021), and Dagmawe et al. (2022), who found that distance from the center of the marketplace hurts market participation [40].

Conclusion and Policy implications

This study examined the determinants of onion commercialization and the level of commercialization among smallholder farmers in Ethiopia, particularly in the Wolayita and Gomo zones of South Region of Ethiopia. The study used both descriptive statistics to summarize household characteristics and an econometric model, namely the double-hurdle model, to uncover several factors influencing market participation and the level of commercialization in the onion market. The results of the study showed that the mean level of commercialization of smallholder farmers in the study area was 79.51%, which indicates that on average households are mainly producing for commercial purpose.

The findings revealed that demographic, socioeconomic, and institutional factors play significant roles in determining smallholder farmers' participation and level of commercialization in the onion market. Family size, farming experience, frequency of extension contact, and distance to the nearest market are significant factors affecting the market participation of onion producers. Educational level, family size, frequency of extension contact, land size allotted for onion production, and distance to the nearest market were factors affecting the level of onion commercialization.

Thus, the following recommendations were made.

Efforts should be made to improve transportation infrastructure, although investing in better road networks and transportation facilities can lower transportation costs and increase the level of commercialization. Furthermore, improving access to education and promoting agricultural training programs can empower farmers with the necessary knowledge and skills to engage in the market more effectively. Moreover, strengthening extension services and the frequency of extension visits can significantly improve farmers' access to information on modern onion production techniques, market opportunities, and new varieties. This can enhance the productivity and market participation of smallholder farmers. Finally, the continuation of research efforts to explore additional factors influencing onion market participation and commercialization is essential.

Disclosure statement

There is no conflict of interest declared by the author(s).

Data availability

The datasets generated and analyzed during this study are not publicly available to protect the privacy of the respondents. However, they can be obtained from the corresponding author upon reasonable request.

Acknowledgments

We would like to express our deepest gratitude to the Ethiopian Institute of Agricultural Research for its funding, which made this research possible. Our heartfelt thanks go to the dedicated data collectors whose hard work and commitment were crucial in gathering the necessary data for this study. We are also immensely grateful to the households who participated in this research. Your cooperation and willingness to share your information were invaluable. This research would not have been possible without the contributions of each and every one of you.

1. Abdu M, Melkamu B, Mohammed A (2016) Smallholder commercialization and commercial farming in coffee-spice based farming system of South West Ethiopia. International Journal of Research Studies in Agricultural Sciences 2: 13-26.

Google Scholar

2. Addisu GT (2018) Determinants of commercialization and market outlet choices of Tef: The case of smallholder farmers in Dendi district of Oromia, Central Ethiopia, A Master thesis, Haramaya University Ethiopia.

Google Scholar

3. Agerie NW, Fikeremaryam BF, Tadie MA, Gebrehiwot A, Wassihun AB, et al. (2020) Analysis of maize commercialization among smallholder farmers: Empirical evidence from north western Ethiopia. Lett Spat Resour Sci 15: 113–127.

Google Scholar

4. Aklilu BM, Merkineh MM, Barana BB (2020) Causes and impacts of deforestation and forest degradation at Duguna Fango Woreda. Int J Nat Resour Ecol Manag 5: 14-25.

Google Scholar

5. Arega DA, Manyong VM, Omanya G, Mignouma HD, Bokanga M, et al. (2008) Smallholder Market Participation under Transactions Costs: Maize Supply and Fertilizer Demand in Kenya. Journal of Food Policy 33: 318-328.

Google Scholar

6. Asfaw SG, Tebarek LM, Meskerem AT (2024) The extent and determinants of vegetable commercialization among smallholders in Sebeta Hawas Woreda, Oromia Region, Ethiopia. Cogent Social Sciences 10: 2299133.

Google Scholar

7. Ater EA, Mutai BK, Bett HK (2021) Factors influencing commercialization of horticultural crops among smallholder farmers in Juba, South Sudan. Journal of Economics and Sustainable Development. 12: 45-56.

Google Scholar

8. Banchamlak H, Akalu T (2022) Vegetable market supply by smallholder farmers in Ethiopia. Cogent Soc Sci 8:1-23.

Indexed at, Google Scholar, Crossref

9. Barrett CB (2008) Smallholder market participation: Concepts and evidence from eastern and southern Africa. Food Policy 33: 299-317.

Google Scholar

10. Berhanu G, Moti J, Hoekstra D (2009) Smallholders, institutional services, and commercial transformation in Ethiopia. Agric Econ 40:773- 87.

Google Scholar

11. Carletto C, Corral P, Guelfi A (2017) Agricultural commercialization and nutrition revisited: Empirical evidence from three African countries. Food Policy 67: 106-118.

Indexed at, Google Scholar, Crossref

12. Cragg J (1971) Some statistical models for limited dependent variables with application to the demand for durable goods. Econometrica 39:829-844.

Google Scholar

13. Direslgne M, Begosew M, Gistane A (2016) House hold dietary diversity and associated factors in Mirab Abaya wereda southern Ethiopia 2016; community based cross sectional study. Diver Equal Health Care 13: 293-296.

Indexed at, Google Scholar, Crossref

14. Dubale A, Fikadu M, Rijalu N (2022) Commercialization level and determinants of market participation of smallholder wheat farmers in northern Ethiopia. African Journal of Science, Technology, Innovation and Development 14: 428-439.

Google Scholar

15. Efa GT, Degye G, Tinsae D, Tadesse K (2016) Determinants of market participation and intensity of marketed surplus of teff producers in Bacho and Dawo districts of Oromia State, Ethiopia. Forthcoming, Journal of Agricultural Economics and Development 5: 020-032.

Google Scholar

16. Engel C, Moffatt PG (2014) Commands to implement double-hurdle regression. The Stata Journal 14:778-797.

Google Scholar

17. Engida G, Agegnehu W, Kusse H (2021) Determinants of sorghum crop commercialization the case of Southwest Ethiopia. Heliyon 7:1-8.

Indexed at, Google Scholar, Crossref

18. Gani BS, Adeoti AI (2011) Analysis of market participation and rural poverty among farmers in northern part of Taraba State, Nigeria. Journal of economics 2:23-36.

Google Scholar

19. Gebreslassie H, Kebede M, Kiros MA (2015) Crop Commercialization and Smallholder Farmers Livelihood in Tigray Region, Ethiopia. Journal of Development and Agricultural Economics 7: 314-322.

Google Scholar

20. Goldman IL (2011) Molecular breeding of healthy vegetables. EMBO reports12: 96-102.

Indexed at, Google Scholar, Crossref

21. Guta RM, Rijalu N, Adugna EB, Diriba BN (2020) Smallholder market participation and its associated factors: Evidence from Ethiopian vegetable producers. Cogent Food & Agriculture 6: 1-15.

Google Scholar

22. Hagos AR, Abraha KZ, Gebreslassie WL, Beyene WG (2018) Assessment of production potential and post-harvest losses of fruits and vegetables in northern region of Ethiopia. Agriculture & Food Security 7: 1-13.

Google Scholar

23. Holloway GJ, Ehui S (2002) Expanding market participation among smallholder livestock producers: A collection of studies employing Gibbs sampling and data from the Ethiopian highlands. 1998-2001 (No. 48).

Google Scholar

24. Ibrahim A, Abduselam F, Assefa A, Alemayehu O, Mulubrihan B, et al. (2021) Determinants of market participation among smallholder vegetable producers in Southwest Ethiopia. Ethiop J Appl Sci Technol 12: 24-37.

Google Scholar

25. Komarek A (2010) Determinants of banana market commercialization in Western Uganda. Afr J Agric Res 5:775-784.

Google Scholar

26. Kusse H, Engida G, Agegnehu W (2022) Determinants of market participation among smallholder farmers in Southwest Ethiopia: double-hurdle model approach. Agriculture & Food Security 11: 18.

Google Scholar

27. Muluneh BE, Mosissa CA, Bikila OF (2019) Major onion (Allium cepa L.) production challenges in Ethiopia: A review. Journal of Biology, Agriculture and Healthcare 9: 42-47.

Indexed at, Google Scholar Crossref

28. Musah AB, Bonsu OAY, Seini W (2014) Market participation of smallholder maize farmers in the upper west region of Ghana. African journal of agricultural research 9: 2427-2435.

Google Scholar

29. Newman C (2001) The infrequency of purchase and double hurdle model of Irish households‟ meat expenditure. Eur Rev Agric Econ 28:393-412.

Indexed at, Google Scholar, Crossref

30. Nigus G, Tsegaye M (2022) Analysing the Factors that Influence Market Participation among Avocado Producers in Kaffa Zone of South-Western Ethiopia. International Journal of Fruit Science 22: 794-808.

Google Scholar

31. Ogutu SO, Qaim M (2019) Commercialization of the small farm sector and multidimensional poverty. World Development 114: 281-293.

Indexed at, Google Scholar, Crossref

32. Olwande J, Smale M, Mathenge MK, Place F, Mithöfer D, et al. (2015) Agricultural marketing by smallholders in Kenya: A comparison of maize, kale and dairy. Food Policy 52:22-32.

Google Scholar

33. Osmani AG, Hossain E (2015) Market participation decision of smallholder farmers and its determinants in Bangladesh. Економика пољопривреде 62:163-179.

Indexed at, Google Scholar, Crossref

34. Tadele M, Wudineh G, Agajie T, Ali C, Tesfaye S, et al. (2017) Analysis of wheat commercialization in Ethiopia: The case of SARD-SC wheat project innovation platform sites. African Journal of Agricultural Research 12: 841-849.

Google Scholar

35. Tamrat G, Mengistu K, Arega S, Getachew L (2023) Drivers of crop commercialization in central and southern Oromia, Ethiopia. Discover Food 3: 15.

Google Scholar

36. Taye M, Abebe BD, Tadie MA (2018) Determinants of commercialization by smallholder onion farmers in Fogera district, South Gondar Zone, Amhara national regional State, Ethiopia. Journal of Development and Agricultural Economics 10:339-351.

Google Scholar

37. Taye M, Abebe BD, Tadie MA (2018) Determinants of commercialization by smallholder onion farmers in Fogera district, South Gondar Zone, Amhara national regional State, Ethiopia. Journal of Development and Agricultural Economics 10:339-351.

Google Scholar

38. Teklebrhan A, Bezabih E, Girma GG (2020) Factors affecting onion market supply in the Medebay Zana district, Tigray regional state, Northern Ethiopia. Cogent Food Agric 6:1712144.

Indexed at, Google Scholar, Crossref

39. Zelalem F, Amsalu M, Guta RM (2023) Commercialization of teff growers and determinants in west Ethiopia: double hurdle model analysis. Journal of Agriculture and Environment for International Development 117: 141-162.

Google Scholar

40. Zelalem F, Amsalu M, Guta RM (2023) Commercialization of teff growers and determinants in west Ethiopia: Double hurdle model analysis. Journal of Agriculture and Environment for International Development (JAEID) 117: 141-162.

Google Scholar