The study was conducted in Mbeya region which is located at the latitude −8˚29'59.99" S and longitude 33˚00'0.00" E and Songwe region which is located at the latitude 8˚31'25.68"S and longitude 32˚32'14.28"E. The three districts of Mbeya, Rungwe and Busokelo were selected from Mbeya region as well as Mbozi and Ileje districts of Songwe region. The study purposively selected the two regions since are among potential regions for avocado productions in Tanzania and are the areas where KIBOWAVI project was implemented ( Figure 1 ).

The study used a multistage sampling procedure which involved the use of purposive selection procedures in selecting regions (Mbeya and Songwe) and five districts (Mbeya, Rungwe, Busokelo, Ileje and Mbozi). The selection of the districts was based on the fact that the selected areas are major producers of avocado fruits in these regions. Then, simple random selection procedure was used to select 209 avocado producing households. Rungwe district accounted for 54% of the study’s sample due to high concentrations of avocado farmers and large number of beneficiaries of KIBOWAVI project which supported horticultural farmers. Additionally a total of 15 key informants were selected basing on their basic roles performed on avocado value chains in the study areas ( Table 1 ).

A field survey was conducted involving random selection procedures to determine 209 households from selected five districts while purposive selection procedures were used to select fifteen (15) key informants 5 LSP, 5 traders/brokers and 5 DEOs. Well-structured questionnaires with both open and closed ended questions were used to collect data from avocado growing farmers. Structured interviews were used to collect data from selected key informants.

Collected data were cleaned, coded and analyzed using Statistical Package for Social Science (SPSS Version 25) with an aid of analytical frameworks. Descriptive statistics were used to characterize collected data. On the othehand, the study employed inferential statistics like linear regression model which was used to determine factors that significantly associated with market performance of avocado fruits among smallholder farmers in the study areas. The linear regression model was specified on Equation (1) and model variable descriptions and expected signs shown in Table 2 ;

$Y={\beta }_0+{\beta }_1{x}_1+{\beta }_2{x}_2+{\beta }_3{x}_3+{\beta }_4{x}_4+\cdots +{\beta }_n{x}_n+{\gamma }_1{\omega }_1+{\gamma }_2{\omega }_2+{\gamma }_3{\omega }_3+\cdots +{\gamma }_n{\omega }_n+\epsilon$ (1)

Additionally, the study used Gross Margin (GM) to analyze profitability in avocado crop cultivations. The model was specified on Equation (2);

$\text{Gross Margin}=\text{Total Revenue}\left(\text{TR}\right)-\text{Total Variable Costs}\left(\text{TVC}\right)$ (2)

Where:

$\text{Total Revenue}\left(\text{TR}\right)$ = Avocado selling Price $\left(P_y\right)$ x Quantity of avocado produced $\left(Y\right)$.

$\text{Total Variable Costs}\left(\text{TVC}\right)$ = Price of inputs $(P_{\omega i})$ x Quantity of inputs used $\left({\omega }_i\right)$

$GM=TR-TVC$

$\begin{array}{c}GM=\left[\text{Avocado fruits selling Price}\left(P_y\right)\times \text{Quantity of avocado fruits sold}\left(Y\right)\right]\\ -\left[\text{Price of inputs}\left(P_{\omega }\right)\times \text{Quantity of inputs used}\left(\omega \right)\right]\end{array}$

$GrossMargin\left(GM\right)=P_{y}Y-P_{\omega i}{\omega }_i$

The findings showed that 88.5% of respondents were male headed households while 7.2% of respondents were female headed households ( Table 3 ). On the other hand, 9.1% of respondents were wives of the heads of the households while 2.4% were husbands of the heads of the households and 0.5% was a male labour working on the farm of household head. The results portray that women involvements in avocado farming is lower than men as was also reported by [1] . But these findings differs with [11] who reported that in United State of America women outnumbers men in avocado productions. Also, differs with what was reported by [12] who showed that in Ethiopia avocado productions is dominated by women as it is grown in homestead giving them high chances to own the crop. Majority of avocado farmers 61.2% had the age between 36 to 39 years, 30.1% had the age between 60 years and above and 8.6% had the age below 35 years. This is an indication that avocado production is highly dominated by elders who own resources needed for investment in avocado productions including fixed assets like land. This was also reported by [13] who showed that the mean age of avocado farmers was 44.71 years. Also, it was found that majority of avocado producers were married (84.7%), widows (9.1%), single (3.3%) and separated (2.9%).

Results suggest that majority of farmers who invest on avocado crop are settled which gives them highest chances of supervising all activities in their fields. Findings showed that 74.7% of avocado farmers had primary education level, 10.5% had secondary education level, 9.1% had informal education level, 3.8% had college education level and 1.9% had university education level. This implies that most of farmers have basic education level which is in line with [14] . Findings showed that the average owned land size for the avocado farmers was 4.3 acres while areas cultivated with avocado was 1.8 acres as was reported by [7] .

Based on the study findings, 90.9% of avocado farmers grows improved varieties mainly Hass while 6.7% and 2.4% grows local and both improved and local varieties respectively ( Figure 2 ). This revealed a rapid transformations on improved avocado varieties especially Hass due to the increase in demand of avocado fruits for exports. Hass variety is highly cultivated due to low gestation periods, and highly demanded for both local and international markets. These findings aligns with [15] and [4] who altogether reported rapid increase in demands of avocado fruits in the cross border trade that has tremendously increased the cultivations of improved cultivars mainly Hass variety.

The results show that majority of farmers sell their avocado fruits at the farm gate accounting for 88.5% which is an indication that farmers end up with produces shares ( Table 4 ). Other market channels used by farmers to market their avocado fruits are; 7.2% sold at village markets, 1.9% sold to the markets outside the village, 0.5% sold direct to the buying and processing factories, 1.4% sold to the neighbors at home and 1.9% sold at the collection centres. These market channels align with [10] who showed that avocado marketing channels differs but majority of farmers sells their produces at the farm gate ending up with relatively low prices.

Numbers in parenthesis are percentages (%).

Processors were found to dominate avocado market (39.7%) in Rungwe, Mbozi, Ileje and Mbeya districts. Processors buy avocado fruits through local collectors and brokers and farmers groups/ cooperatives. On the other hand, 34.9% of respondents reported that the main buyers are local merchants, 25.4% reported to sell their avocado fruits to middlemen’s, 16.8% sells their produce to merchants from other districts or regions, 4.8% sells their avocado fruits to neighbors and 1.9% sells their produce to local collectors ( Table 5 ).

Numbers in parentheses are percentages (%).

The diversity of avocado buyers together with lack of well-structured guidelines and low bargaining power of farmers tend to govern avocado marketing and affect the market performance.

1) Avocado farmers’ membership to associations

The study showed that, most of avocado producers comes from Rungwe district if compared to other districts ( Figure 3 ). However, the avocado producers in Mbozi, Rungwe, Ileje, Mbeya and Busokelo districts have divided into two main groups; the first group are members of the farmer groups and those who are not in farmers’ groups. Most of the avocado growers 34.9% were member of the association and all from Rungwe district, followed by 18.7% of farmers who are not in farmers’ groups. Furthermore, Busokelo and Mbozi districts have the same proportion of farmer groups/associations who produce avocado fruits 10.5%, while others who are not in farmer groups are 6.2% and 2.0% for Mbozi and Busokelo districts respectively. The situation has been different for Ileje district, where the avocado producers who are not in farmer groups/associations were 5.7% and those in farmer groups were 5.3%. In Mbeya district, the proportions of avocado growers were 3.8% and 2.4% for farmer groups/associations and non-members respectively.

2) Service received from associations

The findings revealed that, 53.6% of avocado farmers benefit in terms of marketing services from their associations through collection, storage, collective bargaining and marketing information ( Table 6 ). Results showed that, 26.8% of avocado growers benefit in terms of extension advisory services and 1.9% benefits in terms of inputs supply from their associations. The agricultural services offered by avocado farmers associations have been pointed out and marked to reduce the operational costs while increasing profit margin through strengthened bargaining power.

Numbers in parenthesis are percentages (%)

1) Selling prices

Despite of the facts that most of smallholder farmers in Tanzania sells their avocado fruits at the lowest price, findings revealed that there was significance difference in prices of avocado fruits among districts. The price of avocado fruits with good quality was observed to range from TZS 1,100 to 2,000 per kg. Mbozi district was found to have the highest market price TZS 2,000 compared to Mbeya district with the lowest price of TZS 1,100. Busokelo and Rungwe had the price of TZS 1,400, while Ileje had the price of TZS 1,600 per kg. The study observed that districts with strong managements of cooperatives significantly sold their avocado fruits at higher prices.

With regard to low quality of avocado fruits, the selling prices was observed to be TZS 500, 400, 300, 500 and 400 per kg in Mbeya, Busokelo, Rungwe, Mbozi and Ileje districts respectively ( Figure 4 ). Data showed that there was no great variations in selling prices of rejected avocado fruits across districts. But, unexpectedly the study found that brokers/ middlemen uses advantages of low buying prices for avocado fruits with low quality “Rejects” to raise their profit margin by rejecting avocado fruits with good quality so as to buy them as rejects. Also, inefficiency of farmers’ cooperatives and lack of avocado storage facilities led to low prices. Findings showed no significant difference on market price for avocado to members and nonmembers of farmers’ cooperatives. But, farmers who are members benefits through reductions of production and marketing costs.

2) Profitability of avocado farming

To understand the market performance of avocado fruits in the study areas, the study used profit margin as a proxy implying that the higher the price and low costs of marketing avocado as recommended by [14] . Gross margin analysis showed that there was a significant differences on profit margin of avocado farming across districts. Mbozi district had the highest gross margin of TZS 1,583,022.4 per acre followed by Mbeya district with gross margin of TZS 1,140,707.7 per acre, Busokelo district with TZS 890,230.8 per acre, Ileje district with TZS 333,400.0 per acre and Rungwe district with the least gross margin of TZS 232,933.3 per acre ( Table 7 ). Low profit margin per acre in Busokelo and Rungwe was attributed by low selling prices and high marketing costs. This is also linked with low number of farmers’ memberships in farmers groups/ cooperatives that weakens their bargaining powers which is in line with [6] .

3) Percentage returns on investments

To understand the rate of returns on investments the study estimated the percentage of returns on investing in avocado productions. With the study’s findings it was observed that percentage rate of returns on investments differed across districts whereby Mbeya district had the highest returns on investment of 508.2%, Mbozi district with 367.7%, Busokelo district with 219.9%, Ileje district with 98.7% and Rungwe district had the lowest returns on investment of 39.0% ( Figure 5 ). Low rate of returns on investments for Rungwe district was due to low selling prices and productivity while the marketing costs were high, the Ileje district had comparative low rate of returns due to high marketing costs.

The findings showed that 37.3% of farmers faces the challenges of price fluctuations ( Table 8 ). Unstable price in the study areas was caused by lack of institutional support to set and regulate the selling prices of avocado fruits. Therefore, buyers mainly brokers and middlemen takes the advantage by setting lowest buying prices. Unreliable markets of avocado fruits 32.5% was also a challenge that faces farmers in all of the study districts. Avocado is a perennial crop whose trade is conducted seasonally. Therefore, most of avocado grower’s faces unreliable market issues when avocado fruits matures but they find no place or buyers. Furthermore, results showed that about 8.1% of avocado growers faces difficulties in grading avocado fruits, the study also showed that 5.3%, 4.3% and 3.8% of avocado farmers faces challenges on delayed payment, high input prices and difficulties in transporting avocado fruits respectively. Although, the study revealed that, 4.8% of avocado growers had no challenges when marketing their avocado fruits.

Numbers in parenthesis are percentages (%).

The study showed that, when holding other factors constant; farmer’s experience in avocado farming was statistical significant predictor of selling prices at P < 0.05 ( Table 9 ). This implied that with additional increase in one year of experience in avocado farming increased the selling price by TZS 11.987 per kg. Size of the farm under avocado farming was also found to be a good predictor of avocado selling prices at P < 0.05, such that with additional increase in land size by one (1) acre increased selling prices by TZS 25.401 per kg. Location of the farmer was also significant at P < 0.05 such that a farmer being in Mbeya district had less chance of getting additional TZS 242.839 per kg than farmers in other locations (Mbozi, Rungwe, Busokelo and Ileje). Also, variety grown by avocado farmers was statistical significant at P < 0.05 such that growing local varieties significant reduced market price by TZS 231.380 than improved varieties. Gender of avocado growers and education level were found to be statistical insignificant at P < 0.05.

*** P < 0.01, ** P < 0.05 and * P < 0.1.

This study was designed to evaluate market performance of the avocado fruits in Mbeya and Songwe regions. The socio economic characteristics of respondents showed that avocado productions in Mbeya and Songwe regions is highly dominated by male which has been reported by other studies as reported by [15] . This is due to the facts that most of the men in these regions are considered as heads of households who owns basic resources vital for avocado productions. Married farmers seemed to outnumber other marital statuses of avocado farmers. This implies that with regards to the nature of the crop, it needs farmer who have settled to have a close supervision that may help to realize returns on investments. In terms of education levels, results showed that avocado productions doesn’t need only literate farmers rather it needs commitments. This was evidenced by high representations of avocado growers with primary education level which is termed as basic education level in the country. Such findings aligns with those reported by [5] . The study found that majority of smallholder farmers grow avocado in small pieces of lands and in some occasions they intercrop with some friendly tree species. Also it was found that some farmers intercrop avocado trees with other crops like banana, coffee and other staple crops like maize and beans. This findings aligns with many studies which reported that avocado productions in Tanzania is highly dominated by smallholder farmers who owns less than five acres.

In the study areas, majority of farmers cultivates improved avocado cultivars/varieties mainly Hass. This is highly linked with the rapid increase in demand of improved avocado fruits for exports as shown by [13] . However, it was noticed that prices of avocado fruits greatly varies across districts despite the facts that the varieties, qualities, and all attributes required by the buyers are the same. Mbozi district showed an outstanding market performance of avocado fruits while Mbeya was the district with the lowest market price, the reason explained to be early maturity before the market is flooded by produce from other areas of the Southern Highlands as was found by [5] . But across districts there were slightly differences in terms of prices between members and non-members for farmers’ groups/associations. Low market prices and low productivity were found to be among the reasons for low profitability in avocado farming. Ileje, Rungwe and Busokelo were the districts with the lowest yield per acre while Mbeya was the district with the least market price while farmers in Mbozi districts sold their avocado fruits with the highest prices. Also, results from gross margin analysis showed that, there were differences in level of profit across selected districts. Mbozi district was the one with highest profit margin attributed by high market price. This was due to low average costs and high yield per acre as reported by [7] and [13] . Rungwe district was found to have lowest profit margin compared to all other districts in the study areas.

The findings from linear regression analysis showed that variety grown had significant influence on selling prices. This was due to the fact that the increase in global market demand of avocado fruits has tremendously increased the change from the cultivations of local cultivars to improved cultivars or varieties such as Hass. This aligns with studies by [10] and [16] who showed that production of improved avocado varieties significantly increase the selling prices of avocado fruits in both local and international markets. The experience of a farmer in production of avocado fruits significantly influenced seeing prices which is associated with having the connections with previous buyers and better understandings on the trends of the avocado selling prices. The experienced farmers were also found to have strong abilities to bargain for harvested avocadoes compared to the new farmers. This is in line with a study by [12] conducted in Ethiopia who reported that farmer experience had higher chances of increasing production and fetching higher prices for avocado fruits. It also aligns with results of a study conducted in Mexico by [17] who reported that farmers experience influences the selection of market channel which significantly influences market prices. Location of the farmer was found to be a significant factor influencing market price associated with socio-economic factors of such location. A farmer who is located far from the market place was found to have low market prices while for the locations which were near to markets had relative higher prices. Additionally, location of avocado farmer differentiated the quality of avocado due to weather factors which also influenced the market prices. This is in line with a study by [18] who showed that a distance from district to the market significantly influences market prices. Moreover, the study found that area under avocado production had significant influence on avocado buyers as they would prefer a farmer who can supply large volume of avocado fruits from a single farm. This reduces costs associated with aggregations of fruits from different and distant farmers. Therefore, farmers with large farm sizes were found to have slightly higher prices than those with small farms as also reported by [19] .

Avocado fruits are considered as green gold to smallholder farmers providing income and employment opportunities in Mbeya and Songwe regions. Despite the remarkable contributions to livelihood and food security of the farmers yet it is highly fragile with regard to market prices. The study’s findings showed that avocado production is dominated by elderly men who almost half of them produce independently and the other half operates in farmers groups/ cooperatives. The market system of avocado fruits in the study areas is not yet well structured as there are no established market standards and selling prices which may be in favor of avocado producers. Regression analysis signifies that varieties grown, farmer location, farm size and experience are the main factors that influence avocado selling prices among smallholder farmers. This implies that these factors may be taken into account in developing market systems that will govern avocado fruits marketing in the study areas.

With regards to these findings it is recommended to establish a price floor for avocado that will be useful and profitable to farmers. The study recommends avocado growers to be motivated to join farmer cooperatives in order to increase bargaining power in the study areas. There is also a need to create accessibility of marketing infrastructures like cold storage facilities which may be used to store avocado fruits during peak harvesting season characterized with low prices and selling fruits during low supply with high prices. Moreover, the study recommends the need to train smallholder avocado farmers on value additions which may help to increase their incomes through increased profit margins.