INTRODUCTION

Agriculture plays a vital role in the economic development of developing countries. In Vietnam, agriculture accounts for about 13.96% of the gross domestic product (General Statistics Office, 2019). Rice is the main food crop in Vietnam, with a cultivated land area of 7.47 million hectares and an output of 43.45 million tons in 2019 (General Statistics Office, 2020). Vietnam used to import rice, but thanks to the “Doi Moi” policy in 1986, Vietnam became one of the largest rice exporters globally (Khai & Yabe, 2011). This success is due to the Vietnamese government constantly focusing on supporting agricultural development resources, especially rice production.

One of the main drivers of agricultural development in developing countries is agricultural extension services (Anang et al., 2020). Access to extension services contrib-utes to improved farmers' adoption of technology (Asfaw et al., 2012). Besides technology transfer, capital is a critical factor for agricultural development (Ngoc & Kiet, 2019). The Vietnamese government has a policy to support low-interest loans to buy materials and machinery for farmers’ agricultural production. This policy has contributed to promoting the mechanization of agriculture in general and in rice production in particular. The use of more machines in agricultural production contributes to promoting commodity agriculture in the direction of specialization and improving the efficiency of farming operations (Thuy-Luu et al., 2017). Agricultural mechanization helps to reduce congestion in tillage and rice harvesting operations. The adoption of mechanical technology improves farm productivity and reduces production costs in Asia (Pingali, 2007). In Vietnam, agricultural mechanization services are often provided by cooperatives. However, several farmers have actively borrowed capital to buy machines to self-supply agricultural mechanization services in their rice production due to the government's agricultural credit support policy.

Farmers have the credit loan to buy agricultural machinery such as plows, cultivators and combine harvester to produce and provide farming services to the outside. The ineffective use of purchased machinery negatively affects the farmer's income. Many studies evaluate the financial efficiency of agricultural production models in Vietnam (Thanh & Phong, 2014;Thuy & Loc, 2015;Nam & Nghi, 2016;An & Loc, 2017;Dung, 2020). However, there is no study to evaluate the financial efficiency of agricultural mechanization service delivery in rice production by farmers. This is the first study to analyze financial efficiency and the factors affecting the financial performance of providing agricultural services by farmers in Vietnam.

Therefore, the main contributions of the article are as follows: (1) to analyze the financial performance of agricultural mechanization service provision in rice production by farmers; (2) to study the factors affecting the financial efficiency of the provision of agricultural mechanization services in rice production and propose policy suggestions to improve the efficiency of farmers’ operations.

The paper is organized as follows. In Section 2, the Materials and research methods are proposed; Results and discussion are represented in Section 3; Section 4 outlines the conclusions and recommendations.

DATA AND MODEL

The study uses data from the 2014 Vietnam Household Living Standards Survey (VHLSS) to evaluate agricultural mechanization service delivery's financial performance. The General Statistics Office of Vietnam conducts this survey works with the World Bank’s technical assistance. The VHLSS data set includes 9,189 households sampled from 3,063 communes across Vietnam. Households are randomly selected and represented at the national, rural, and urban levels.

In 2014 VHLSS data, 143 farmers are providing agricultural mechanization services in rice production. Among them are farmers who offer two or more farming services. However, the study wants to analyze the performance efficiency of each type of agricultural service. Therefore, if the household provides two kinds of farming services, it will be input into two observations. The results of data processing create a sample with 164 farmers who deliver agricultural services to rice production. Table 1 shows that the services provided by farmers in rice production include four types: plowing and soil preparation (67.1%), rice threshing (29.9%), irrigation (1.8%), pest prevention and control (1.2%).

The descriptive statistical method is applied to explore farmers’ socio-demographic characteristics and the revenue, cost, profitability, and financial efficiency of providing mechanized services in rice production. Financial performance is measured in terms of accounting and market value (Murphy et al., 1996). Profitability is usually used as a measure of financial performance. If the profitability is getting higher, it means that the company is efficient in finance and vice versa. The measures of financial performance often include return on sales (ROS), return on assets (ROA), return on equity (ROE), and price to book ratio (P/B) (Ha et al., 2019). However, due to the limitations of VHLSS data and producing characteristics of farmers, the study uses the ROS index to measure the financial performance of the provision of agricultural mechanization services.

Based on the literature review, studies often use the ordinary least square regression (OLS) to examine the factors that affect financial performance (Ha et al., 2019). Assumptions of OLS estimates are tested to check the suitability of the regression model. The variance inflation factor (VIF) coefficients are smaller than four shows that there is a moderate correlation between the explanatory variables in the model but not noticeable (Appendix: Table A1). Because the highest VIF is equal to 2.2, the standard error of estimated bi would be approximately 1.48 (square root of 2.2) times larger than if the i^th predictor is uncorrelated with other predictors. To check homoscedasticity, the White test is applied. The results indicate that chi squared is 0.1388 which is greater than 0.05 (Appendix: Table A2). Therefore, there is no violation evidence of the homoscedasticity assumption. To determine the normality of a random variable, the Skewness Kurtosis test is applied. The result shows that the p-value is 0.0738, which is greater than 0.05 implying its significance at a 5% level (Appendix: Table A3). Therefore, the residuals in the Skewness Kurtosis test show a normal distribution. The regression model is designed as follows:

Where ROS is the index of return on sales of the provision of agricultural mechanization services in rice production by farmers. The formula for calculating ROS is as follows:

The Age variable is the age of the person in charge of providing agricultural mechanization services of the household. Age negatively affects rice production efficiency in Vietnam (Khai & Yabe, 2011). However, older managers perform better in terms of management than younger ones (Kauko, 2009). The Ethnicity variable is a dummy variable, which takes value one if the farmer is of the Kinh ethnic group and zero if the farmer belongs to other ethnic groups. This variable was included to control the basic disposition arising from ethnic differences. Vietnam has about 54 ethnic groups and the Kinh ethnicity accounts for about 86.2% of the population (General Statistics Office, 2019). The Gender variable is a dummy variable, which equals one if the farmer is a man and zero if the farmer is a woman. In Vietnam, men are the main person in charge of agricultural production. However, some households whose primary responsibility in agricultural production is female. Females are less efficient in farming than males (Mishra et al., 2017).

The Education variable is the number of years attending the school of the main person in charge. A high educational level helps farmers grasp information and easily apply science and technology to production. Therefore, educational attainment might positively impact the efficiency of farmers’ agricultural production (Nam & Nghi, 2016). The Head variable is a dummy variable that takes value one if the main responsible person is the household's head. It takes zero if the person in charge is another member of the family. The head is the owner of the machinery for the provision of agricultural services. The owner, concurrently managing, could positively impact the business's financial performance (Thuy, 2020). The Rice variable is a dummy variable that assumes one if rice production is the household’s main activity and zero if otherwise. Rice farmers are more efficient than farmers who produce various crops (Khai & Yabe, 2011).

The Member variable is the number of working-age household members (15-60 years old). In Vietnam, agriculture production is a labor-intensive industry. However, households that use too much household labor for agricultural activities can cause a waste labor force and negatively affect the financial efficiency of agricultural production (Nam & Nghi, 2016). Instead of participating in agricultural activities with family, some household members can do other jobs that generate higher income. The Credit variable is a dummy variable, which takes value one if the farmers borrow capital to buy materials and machinery for rice production and zero otherwise. The credit might increase financial capacity and contribute to improving farmers’ agricultural profitability (Nam & Nghi, 2016). The Harvesting variable is a dummy variable, which equals one if the farmer provides the rice harvesting service and zero if the farmer provides other agricultural services. Different types of farming services have different rates of profit. Financial performance depends on the production model (Thuy & Loc, 2015). The Active month variable is the number of months the farmer provides agricultural services in one year. Producing too many crops a year leads to decreased crop yields (Cao et al., 2017). However, the increase in the number of crops leads to profit growth in agricultural production (Thuy & Loc, 2015).

The Depreciation variable is dummy, takes one if the agricultural machinery of farmer still has depreciation time and zero otherwise. Depreciation is included in the cost of production. Increasing production costs negatively affects farmers’ efficiency of agricultural production (An & Loc, 2017). The Outsourced labor variable is a dummy variable, equals one if the farmer hires labor force outside the family and zero otherwise. Hiring outsourced labor will increase the production costs of farmers. Besides, when labor reaches the threshold, productivity gradually decreases (Thanh & Phong, 2014). The Cultivation land variable is the cultivated area of rice land each year of the household. The different number of crops leads to the unlike output and profit despite the same land area. Therefore, the annual cultivation rice area variable is used instead of the land area. The land area positively affects agricultural production efficiency (Nam & Nghi, 2016;Cao et al., 2017;Dung, 2020). The Association variable is a dummy variable, which equals one if the farmer joins the Farmer's Association and zero otherwise. Joining the Farmers Association contributes to increasing agricultural productivity (Thanh & Phong, 2014).

β are the parameters to be estimated, and ε is the error of the model. Table 2 represents the variables included in the regression model to examine the factors that affect the financial performance index ROS. The average age of the farmer in the survey sample is 44.6 years. The average number of years of schooling is 7.8 years, and each household has about 3.1 people of working age. The average rice cultivated land area of the family is 1.25 ha.

RESULTS AND DISCUSSION

Characteristics of the farmers

The socio-demographic characteristics of farmers providing agricultural mechanization services are presented in Table 3. Most farmers are male, accounting for over 76%. Farming activities and providing agricultural mechanization services are quite heavy for women, but they can still take the primary charge by planning operations and hiring outside workers. The Kinh ethnicity accounts for 86% of farmers in the surveyed sample. This data is consistent with the Vietnam census data in 2019 (General Statistics Office, 2019). Over 92% of the sample farmers are working age, and 7.9% of the farmers have reached retirement age. However, farmers in Vietnam do not have the concept of retirement age. They only stop working when they are no longer healthy enough for agricultural production. The majority of farmers have secondary education (57.9%), and only 2.4% of the survey sample farmers do not attend school. Most households have from one to four members who are of working age. There are 46.3% of farmers in the survey sample who is head of household. The household heads are not responsible for providing agricultural mechanization services mainly because they have other jobs such as housework and livestock or are not healthy enough to work. The main activity of the households in the sample is agriculture. In which about 58.5% of households have their primary income from rice production. Nearly half of the surveyed farmers participated in the Farmers Association. Survey on credit to buy materials and machinery for rice production, about 24.4% of farmers have the official loan for this purpose.

The financial performance of agricultural mechanization service delivery

Table 4 represents the revenue of agricultural mechanization service delivery. The average active duration per year is about 2.8 months which depends on the rice production crop. In Vietnam, farmers produce one to three rice crops per year, depending on the locality. Each household providing agricultural mechanization services in rice production only operates for about one month in each crop. A short uptime affects the farmers’ return on investment in agricultural machinery. However, farmers provide agricultural mechanization services in rice production for 12 months per year. These farmers move their machines to rice fields in many different localities to operate. Middleman is required if farmers want to provide services outside their living area. They have to pay a brokerage fee based on the acreage. Earnings per month reach 13.6 million VND. The average revenue from the provision of agricultural mechanization services is 39.4 million VND. There is a significant disparity between the farmer with the highest income (250 million VND) and the lowest (150,000 VND).

Details of the costs of agricultural mechanization services delivery are presented in Table 5. The average cost of providing mechanization services in rice production is about 22.2 million VND, but the standard deviation is 63.3 million VND. Fuel and energy costs are the most significant expenses, accounting for about 50% of the total cost. In this cost, in addition to the energy and fuel for the operation of machines in the rice fields, farmers have to spend fuel costs for moving their devices between apart plots. Rice land in Vietnam is quite fragmented, which negatively impacts farmers’ income (Tran & Vu, 2019). Other enormous costs are asset depreciation, outsourcing, and minor repairs, respectively. The remaining costs account for a negligible proportion.

Table 6 indicates the return on sale index of agricultural mechanization service delivery in rice production. The average ROS index of agricultural mechanization service delivery is 56.9%. It is larger than the ROS index of other farm activities in Vietnam, such as rice cultivation is 36% (Dung, 2020), chili cultivation is 31% (An & Loc, 2017), raising pigs, chicken and ducks are 41%, 36% and 25%, respectively (Nghi, 2017). However, considering the average net profit from the provision of agricultural mechanization services per year is only about 17.2 million VND. This is the net profit since farmers’ provision of agricultural mechanization services in Vietnam is not taxable. However, these costs do not include the cost of household labor. Farmers in Vietnam have the concept of “taking effort for profit” so the family labor cost is often ignored when calculating production costs. It is not profitable for the farmer to include the cost of household labor in the total cost of providing agricultural services.

The financial efficiency of the four types of mechanized services in rice production is presented in Table 7. The two kinds of agricultural mechanization services with the highest revenue and cost are rice threshing, plowing and soil preparation. To provide these two types of agricultural mechanization services, farmers need a sizable investment to purchase combine harvesters and plows. However, not all farmers can afford to buy these machines for rice production. Therefore, most farmers hire rice threshing, plowing and soil preparation services in their rice production. Rice farmers in the same area often sow seeds simultaneously, so the harvest time is the same. This leads to a sudden increase in the demand for rice harvesting services during the harvest season. If not harvested in time, the yield and quality of rice are reduced. As a result, farmers are willing to pay a relatively high rice threshing service compared to other agricultural mechanization services. The performance of the combine harvester is also higher than that of other farm machines. These explain why the rice threshing service brings in the most considerable revenue. Pest prevention and control and irrigation services achieve the highest financial efficiency, respectively. However, the profit from these two types of services is negligible as most farmers can provide these services themselves, so there is no need to hire outside assistance.

Factors affecting the financial performance

Table 8 represents the results of the OLS regression model. The R-squared value is 0.4012, indicating that the model’s independent variables explain 40.12% of the variation of the ROS index. Except for the explanatory variables Gender and Rice, the remaining variables such as Ethnicity, Head, Credit, Outsourced Labor, Depreciation and Cultivation land negatively impact the ROS index of providing agricultural service. The variables such as Age, Education, Member, Harvesting, Active month, Association have no statistical significance in the model.

The Ethnicity variable’s coefficient indicates that the Kinh ethnicity has lower financial performance in providing agricultural mechanization services in rice production than other ethnic groups. This is consistent with the research results of Khai and Yabe (2011). However, the data shows that the scale of agricultural mechanization service delivery of the Kinh ethnicity is twice as large as that of other ethnic groups (Appendix: Table A4). In terms of gender, men get higher financial efficiency than women in agricultural service delivery. Research by Mishra (Mishra et al., 2017) supports this finding that females are less efficient in farming than males. Agriculture is heavy work and a labor-intensive industry, so men have the upper hand.

The household head is inefficient in providing agricultural mechanization services compare with other members. This result is inconsistent with the research result on the companies’ financial performance (Thuy, 2020). The limited management capacity of farmers can explain this. The household head has to handle several household tasks, thus affecting the efficiency of service provision. If it is another member of the household in charge of agricultural mechanization services, the head focuses their expertise on doing this only. Therefore, families need to assign the work responsibilities among household members instead of the head managing everything.

Farmers with the primary income from rice production are efficient in financial performance in providing mechanization services in rice production. Agricultural service delivery is not the primary income source of these farmers. They consider taking a break after finishing their rice production or participating in agricultural mechanization service provision. As a result, these farmers only engage in agricultural service delivery when profits meet their expected level. Farmers who get credit or invest in new agricultural machinery have less financial efficiency than other farmers. Because loan interest and asset depreciation are also included in the farmer’s cost of providing agricultural mechanization services, they reduce profitability and financial performance. Therefore, the calculation of the benefit-cost of the loan to invest in new agricultural machinery is necessary.

The financial performance of agricultural mechanization services reduces when using outsourced labor. The cost of outsourcing labor accounts for a large part of the total cost. When the family labor force is insufficient, the farmer must hire workers from outside. Due to labor migration from rural to urban areas, the rural labor source is scarce, and labor hiring price increases. However, migration can reduce the productivity of rural labor (Nguyen et al., 2019). When using labor reaches the threshold, labor productivity decreases (Thanh & Phong, 2014). Therefore, labor resources need to be calculated before investing in machinery providing agricultural services. The research result also indicates that the financial efficiency of agricultural service provision in rice production decreases when farmers’ cultivated rice land area increases. However, an increase of 1,000 m² of paddy land only reduces the ROS index of agricultural service provision of farmers by 0.2%.

CONCLUSIONS

The study analyzes the financial efficiency of agricultural mechanization service provision in rice production of 164 farmers extracted from the 2014 VHLSS data. Agricultural mechanization services in rice production provided by farmers include plowing, rice threshing, irrigation, and pest control. The average annual duration of agricultural mechanization service provision is 2.8 months. The average profit is 17 million VND/year. The most profitable services are rice threshing and land plowing, respectively. The ROS index of providing agricultural mechanization services is about 56.4%. Although the financial efficiency of providing agricultural mechanization services is higher than producing other agricultural products, it is relatively small in terms of the scale of operation. The regression results indicate that factors such as the Kinh ethnicity, person main in charge, agricultural credit, asset depreciation, use of outsourced labor, and rice cultivation area negatively impact farmers' financial performance. Meanwhile, the farmer is male or whose main activity is rice production with high financial efficiency in providing agricultural mechanization services.

The empirical results demonstrate that farmers in Vietnam can achieve high financial efficiency in income-generating activities, in this case, providing agricultural mechanization service in rice production. However, it is necessary to assign the main person in charge of delivering agricultural services instead of the head to cover all household affairs. Credit loans to buy new agricultural machinery need to be carefully calculated in terms of cost-benefit because the machines' short time of operation negatively affects the payback time. Besides, hiring and arranging jobs for outside workers appropriately improves farmers' financial efficiency in providing agricultural services.

Policies to support agriculture in general and credit policies for agriculture have contributed to promoting Vietnam’s agricultural mechanization. The financial efficiency of agricultural mechanization service provision is higher than other agricultural activities, contributing to increased farmers’ income. The government’s policies to support ethnic minorities in Vietnam have contributed to improving their production activities and earnings. However, the production scale of ethnic minorities is still smaller than that of the Kinh ethnicity. Therefore, the Vietnamese government should continue to maintain policies to support ethnic minorities to reduce the income gap.

The limitation of the study is that the analyzed data only included 164 farmers who provided agricultural services in rice production in Vietnam. Because the data is only a tiny part of the 2014 VHLSS, there is a lack of detailed information on providing agricultural services to farmers. Therefore, to get a more general overview of the farming service provision of Vietnamese farmers, it is necessary to have a separate survey of farmers who provide agricultural services. Future researches are essential to interview more detailed information such as the source and amount of investment money in agricultural machinery, time and area of operation in each season, the number of family workers involved in farming service provision. From there, it is possible to analyze farmers’ investment efficiency to provide agricultural services and make more specific and precise strategies for farmers and policymakers.

적 요