ABSTRACT
The objective of the study is to analyze the influence of land fragmentation on the income of rice-producing households. The study collected information from 241 rice-producing households in two communes including Thuy Thanh and Thuy Phu in Huong Thuy town, Thua Thien Hue province. The study used Simpson's diversification index to measure land fragmentation in the study area. We applied the instrumental variable method (2-stage regression - 2sls) to overcome the endogeneity problem that may exist in the model. Research results show that the increase in land fragmentation for rice production has the effect of reducing the income of producing households. From the research results, we propose to strengthen the policy “consolidation and exchange of plots'' as a basis for land accumulation, investment in mechanization to create favorable conditions for households to produce rice on a large scale and increase the application of scientific and technical advances to production to increase income.
Keywords: Land fragmentation, rice income, instrument variable method, Thua Thien Hue
INTRODUCTION
Land is an important resource for socio-economic development activities, especially agricultural production (Phan et al., 2022). Like many countries in the world, the long-term development of Vietnam's agriculture depends heavily on the efficient use of land resources (Marsh et al., 2007). It can be seen that the current land productivity in our country is still quite low compared to other countries in the region and the world (Thu, 2021). One of the causes of that problem is the impact of land fragmentation on agricultural production (Cuc, 2017). Land fragmentation is still a prominent issue in many developing countries, including Vietnam. Rapid rural population growth and policy for equitable land access of farmer households makes the average agricultural land size of households getting smaller and smaller. Currently, our country is one of the countries with a relatively high level of land fragmentation. Specifically, Vietnam has 14.5 million farmers with nearly 70 million plots of fields, an average of 300 - 499 m2, the average household has 7 – 10 pieces, and varies greatly between regions (Son, 2018).
Huong Thuy is a town with a tradition of rice farming, located in the south of Hue city, Thua Thien Hue province. In recent years, the agricultural production policy following the model of "Small farm, Large field" (Anh, P., & Thanh, T. (2014) has received a lot of attention from the people as well as the management agencies. This policy is promoting different small farms having parcels in the same field to use the same variety and to practice the same technical protocol in order to access to mechanical services. Due to the natural, historical, and socio-economic conditions of the locality, the rice land is still fragmented and dispersed, so it is difficult to form concentrated rice production areas to apply advanced technics for high yield. This situation has affected the production efficiency of farmers because it is difficult for farmers to effectively apply mechanization and advanced agricultural technologies. Currently, some localities in Huong Thuy town are encouraging farmers to clear their fields, creating favorable conditions to apply mechanization in production, reduce labor costs, and limit pests and diseases in the production process. (Anh & Thanh, 2014). Such policies are expected to promote land accumulation and mass production of goods and create greater added value for farmers.
The literature review shows that there is a link between the current state of land fragmentation and the outputs obtained from production activities and the livelihoods of households. Land fragmentation can still have positive effects for households, for example by reducing vulnerability to the impact of climate change (Ntihinyurwa & associates, 2019). In addition, the study of Blarel and associates (Blarel, B., Hazell, P., Place, F. & Quiggin, 1992) concluded that reducing fragmentation could actually make farmers worse off. Similarly, research by Kawasaki (K. Kawasaki, 2010) shows that fragmentation reduces production risk for households. However, many studies have also shown the negative aspect of mass fragmentation to the welfare of producing households. Fragmentation can prevent households from applying new technologies in the production process (Tran and Vu, 2019a), requiring more production labor that will increase the cost of production (Ciaian et al., 2018). Some studies also show that fragmentation will make households unable to ensure food security for them (Tran & Vu, 2021; Hung et al., 2007).
As such, land fragmentation can have either a positive or a negative impact on household income, depending on the particular circumstances under consideration (Tran & Vu, 2019a). Therefore, empirical studies are needed to examine the effects of fragmentation on household income in each specific location. Therefore, it is necessary to conduct a study on the effects of rice land fragmentation on household income to provide sufficient information on the positive or negative effects of land fragmentation to the income of rice-producing households in Huong Thuy town. Research results can be the basis for developing local agricultural land management policies. In addition, the results of the study will also contribute to the theoretical and practical overview of the relationship between land fragmentation for rice production and household income.
Literature review
Many previous studies have shown that land fragmentation is a constraint to agricultural activities because fragmentation is a barrier to the use of modern, mechanized equipment, such as tractors, tillers, and harvesters (Tran & Vu, 2019a). In addition, fragmented land may impede the use of more profitable crops on a larger scale (Markussen et al., n.d.). Land fragmentation often requires more labor input for the operation of mechanized equipment, especially the considerable time spent moving between plots (Ciaian et al., 2018). Therefore, fragmented land has a significant negative impact on production efficiency and agricultural growth in many regions such as South Asia, Southeast Asia, etc. (Niroula & Thapa, 2007), (T. Kawasaki & associates, 2020). Research by Quang and Vu (Tran & Vu, 2021) found that Land fragmentation, as measured by the Simpson index, has been found to increase the likelihood that households will be food insecure during the year and have more days without protein-rich foods within a week. Such negative impacts only occur for ethnic minority households, not Kinh/Hoa households. Besides, the study of Hung et al. (Van Hung& et al., 2007) argues that land fragmentation has a negative impact on crop productivity, increased use of family labor and other monetary costs.
Although the effect of land fragmentation has been found to be negative in many studies, several other studies have shown benefits associated with land fragmentation for farm households in some cases (Ntihinyurwa & Vries, 2021). Scattered farms may be more fertile and less susceptible to disease or natural disasters (Markussen et al., 2011). Land fragmentation may reflect farmers holding multiple plots of land of varying quality, allowing them to diversify crops, allocate labor demand, and reduce production and price risks (Ciaian et al., 2018; Ntihinyurwa & Vries, 2021). A study by Ciaian et al., (2018) shows that fragmented land in rural Albania has significantly promoted agricultural diversification, thereby improving food security, and its influences on subsistence farmers are larger than on market-oriented households. Similar results are found for farm households in Rwanda, where land fragmentation has a positive impact on food quality, food sustainability and food security (Ntihinyurwa & Vries, 2021).
The literature review shows that land fragmentation can either benefit or inhibit agricultural production (Markussen et al., 2011), (Ciaian et al., 2018); but there is little evidence of an effect on household income. Land fragmentation can have a positive or negative effect on household income, depending on whether it creates a positive or negative impact on farming efficiency. On the other hand, in certain cases, the disadvantages or costs of land fragmentation may persuade farmers to diversify their livelihoods towards non-agriculture, which can yield higher returns compared to farming (Ntihinyurwa & Vries, 2021). As such, land fragmentation can have a positive or negative impact on household income, depending on the particular case considered. Therefore, a study to assess the impact of land fragmentation of rice-producing households on household income is needed.
To estimate the effect of agricultural land fragmentation on the welfare of households, many previous studies have used different models. Quang and Vu's study (Tran & Vu, 2021) used a logit model to estimate the influence of land fragmentation and other factors on food security of producing households. This can be done efficiently when the dependent variable (household food security) is a binary variable. Similarly, research rescue by Phan et al. also used an ordered probit model to estimate the influence of land fragmentation and other factors on the level of food security of producer households (Phan et al., 2022).
RESEARCH METHODS
The method of data collection
Secondary data collection: Data and information are collected from reports of the town's People's Committee over the years, the town's statistical yearbook, from various departments and local authorities, in addition, secondary data is collected through library, internet, television, and websites.
Primary data collection: The study conducted to collect data on rice growing households in 2 localities including Thuy Thanh and Thuy Phu communes. The two areas represent localities with a large total area in Huong Thuy town. The study conducted a survey of 121 farming households in Thuy Thanh commune and 120 households in Thuy commune based on a convenience sampling method.
Methods of data analysis
The study uses Simpson index to estimate land fragmentation of rice producing households in the study area: Simpson index = 
. In which, the Simpson index is estimated based on the number of cells (n), the size of the cells (a) with the unit m2 and farm size (A) in m2. The Simpson index ranges from 0 to 1, with a larger index referring to more fragmentation.
To analyze the effect of fragmentation on household productivity, the study uses two methods, including linear regression (OLS) and instrumental variable method - 2-stage regression (2sls) to estimate the impact of fragmentation on earnings from manufacturing operations. The proposed theoretical model is as follows:
variable income of households from rice production in 1 crop (VND1,000 /household/crop) in logarithmic form; Simpsoni is the fragmentation coefficient of land for rice production, Zi are control variables such as household characteristics, εi is the error. Based on previous studies (Knippenberg et al., 2018), (Chu et al., 2021), (Niroula & Thapa, 2007), (Tran & Vu, 2019a), variables selected and used in this study include: Age of household head (Year), Education level of household head (Year), Number of household workers (Number of people), Number of working days in rice production of the family (Number of working days) ), Yield (ton), Yield (ton/ha), Gender of household head (1: Male; 0: Female), Rice varieties used (1: New variety; 0: Traditional variety), and Access credit (1: Yes; 0: No).
The objective of the study was to determine the effect of land fragmentation on the income of rice-producing households. In estimating this relationship, the problem of endogeneity between variables may arise because land fragmentation is an explanatory variable but is determined together with household income from rice production along with other factors such as production characteristics of the household. Therefore, the linear regression method can yield skewed and inconsistent results. The instrumental variable method is recommended to be used instead to give more reliable results (Tran and Vu, 2019b). This study estimates and compares the results of both the OLS method and the instrumental variable method.
To use the instrumental variable method estimation, the study uses the instrumental variable as a dummy variable such as the location of the plots ( 
) (Tran & Vu, 2019a) (D=1: household plots in Thuy Thanh commune, 0: household plots in Thuy Phu commune). The instrumental variable method includes two steps in the regression process. In the first step, the method estimates the association of factors belonging to households and the land fragmentation index of households. The method then estimates the effect of land fragmentation on household income from rice production. Regression model phase (2sls) are as follows:
Same as above: 
is the fragmentation coefficient of rice-producing land – representing the endogenous variable, Zi are control variables such as household characteristics, ε1i, ε2i is the error Equation (2) and Equation (3) respectively. The conditions of IV are expressed through the following system of equations:
Thus, the instrumental variable D must satisfy two conditions: (i) D is not correlated with 
; (ii) but D is also correlated with the fragmentation coefficient Simpson. To determine whether the problem of endogeneity exists or not, the study used Wu-Hausman statistical test with the hypothesis H0: The model does not exist endogenous phenomenon, and H1: The model exists endogenous phenomenon. If the P-value is less than 5%, this means that there is an endogeneity problem in linear regression (OLS). In addition, the Cragg-Donald Wald test is used to evaluate the strength of the selected instrument variable.
The study also used non-parametric analysis to graph the distribution of income variables in relation to the change of Simpson's land fragmentation coefficient, in five percentiles (0; 0.25; 0.5; 0.75; 1). Non-parametric testing methods can be used quickly, easily, and just as effectively as non-parametric testing methods for data for which the distribution function has not been determined. This method will give an initial view of the relationship between the variables of interest, thereby laying the foundation for more in-depth analytical methods.
RESULTS AND DISCUSSION
Using agricultural land situation in Huong Thuy town, Thua thien Hue province
In Huong Thuy town, the total land area used for agricultural production is 35,464 ha, accounting for 78% of the local natural land area (Table 1). In which, the total area of rice cultivation is 3,064 ha, accounting for 6.74% of the total land area in Huong Thuy town. Besides, the area for other agricultural production accounted for a fairly large proportion with 70.60%. In general, Huong Thuy town has a diversity of forms of land use in agricultural production.
Situation of rice production in Huong Thuy town
As shown in Table 2, in Huong Thuy town, in 2019, the Rice cultivated area of the whole town was 6,154 ha, not much different from the 2018 area of 6,215 ha; but rice production in 2019 is 39,321 tons, an increase of 430 tons compared to 2018. Winter-spring crop in 2018 - 2019, the cultivated area of the whole town is 3,124 hectares, with varieties like HT1, TBR279, of which, the area of rice production accounts for more than 10%. Although the weather at the beginning of the crop was quite unfavorable, heavy rain lasted for many days, but thanks to the initiative of the agricultural sector and the attention of leaders at all levels, farmers not only did not lose their crops but also achieved good results of a bountiful crop, 85% of Huong Thuy's 2019 winter-spring rice area is well developed, the field uniformity is quite high, and the crop is bountiful. In 2018, the total rice production area was 6,215 ha, of which winter-spring rice accounted for 3,124 ha, summer-autumn rice accounted for 3,091 ha. In 2019 compared to 2018 it decreased by 61 hectares, equivalent to 99.03%. Winter-spring rice and summer-autumn rice decreased by 32 hectares and 29 hectares, respectively, accounting for 98.98% and 99.08%. In terms of output, in 2019 compared to 2018, the total output increased by 430 tons, equivalent to 101.10%. Winter-spring rice decreased by 287 tons compared to 2018 of 20,379 tons. Summer-autumn rice fluctuated sharply to 716 tons, equivalent to 103.87%. This figure shows that in the summer-autumn crop in 2019 compared to 2018 there is an improvement in seed quality as well as favorable natural conditions for cultivation. In 2020, the area of all crops decreased due to the influence of many factors affecting the farming process, the total area decreased by 59 hectares compared to 2019 to 6,154 hectares, equivalent to 99.20%. In which, the winter-spring crop decreased by 35 hectares, accounting for 98.86%; Summer-autumn crop reduced 24 hectares accounted for 99.55%. Output in this year also decreased significantly because of the prolonged disease situation, affecting many aspects of farmers, especially the situation of rice trading after harvest, down 1,822 tons for the whole year and decreased by 1,996 tons in the summer-autumn crop. Productivity in 2019 compared to 2018 increased by 1.32 quintals/ha. It's a small number, but it affects the whole production process. But due to the COVID-19 pandemic, the yield in 2020 compared to 2019 decreased by 2.4 quintals/ha.
Basic information about surveyed households
A survey of 241 rice-producing households in two areas shows that the fragmentation coefficient in Thuy Phu commune is lower than that in Thuy Thanh commune, reaching 0.39 and 0.42, respectively (Table 3). The average age of the household head is over 50 years old. It is a traditional field and requires experienced people to produce, so the age of the household head will be quite old. However, the education level of the household head is not high, about 7 to 9 years on average. In the countryside, the main occupation is agriculture, the issue of cultural level is often not paid due attention by the farmer households. What farmers care about is production skills and experience. Besides, the data on the labor force in rice production is about 2 members per household and the number of man-days in rice production is about 71 man-days per household. The average yield of rice-producing households is 6.65 tons/ha. Therefore, there was a similarity in productivity between surveyed households in the 2 study areas (Table 3).
The survey also shows that the majority of household heads in rice production are predominantly male, with over 70% of household heads interviewed being male. A large percentage of households have adopted new rice varieties to mitigate some of the risks in production due to the effects of climate change. This also helps households reduce investment costs to reduce risks in production. The percentage of households using new rice varieties is 67.5% of the total number of households interviewed. Credit also plays an important role in agricultural production. Access to credit helps households have enough financial resources to pay for input production costs. The percentage of households accessing credit accounts for over 90% of the total number of households interviewed.
Impact of land fragmentation on household income
Non-parametric analysis results
The results of the non-parametric analysis are presented in Figure 1, in which the household income from rice cultivation is presented according to the five percentiles of the Simpson coefficient of fragmentation: 0-0.25;0.25-0.5;0.5-0.75 and 0.75-1. The results show that at the low percentile level, corresponding to a low fragmentation coefficient (0-0.25), the income of households from rice cultivation tends to be the highest. This income distribution tends to decrease gradually with higher percentiles such as 0.25-0.5; 0.5-0.75; and 0.75-1. These initial results show that there exists a strong relationship between income and land fragmentation. However, to clarify and find statistical evidence of this relationship, applying more complex statistical methods is necessary and we present in the next step.
Regression Result OLS and 2SLS
The study estimated the impact of productive land fragmentation on household income by two methods: linear regression (OLS) and instrumental variable method. The p-value of the Wu-Hausman test for endogeneity is less than 5%. Therefore, there is an endogenous problem when estimating the impact of land fragmentation on household income. Therefore, estimation by instrumental variable method is necessary. The Cragg-Donald Wald (F) statistic is greater than 10, so the instrumental variable method results are likely to be reliable.
Estimation results of the regression model by means of instrumental variables show that increased land fragmentation has a negative impact on household income and is statistically significant with a coefficient of -2.994 at the level of statistical significance 5% (Table 4). This result is similar to the study of Tran and Vu (Tran & Vu, 2019b). This is explained as when land is divided into smaller pieces, it can be difficult to efficiently use resources such as labor, machinery and capital for production. In addition, fragmented land can lead to difficulties in managing and maintaining efficiency in the production process, thereby reducing income generation potential for households. Land fragmentation is also a factor preventing people and businesses from making long-term, large investments in agriculture.
The reality is that farmers cannot determine the market price of their output, so increasing production can be a way to help producers gain more profits. In addition, if farmers can produce rice on a large scale with low cost, they can increase profit margin, which will also contribute to increasing household income. With the strong development of scientific and technical progress, farmers can increase rice production through the use of new varieties, new techniques, etc. These measures are only really effective when households farming on a large scale to increase the efficiency of input use in the production process. The accumulation of land has contributed to reducing the small-scale production area, creating a large number of agricultural products, high income value and creating favorable conditions for farmers to participate in production and application, use science and technology. Reducing land fragmentation in rice production is an important strategy to develop Vietnam's rice production in the direction of large commodities.
CONCLUSIONS AND RECOMMENDATIONS
The objective of the study is to analyze the effect of land fragmentation on the income of rice production. A survey of 241 rice-producing households in two areas shows that the fragmentation coefficient in Thuy Phu commune is lower than that in Thuy Thanh commune, reaching 0.39 and 0.42, respectively. Two methods are used in the study, including: linear regression (OLS) and instrumental variable method to analyze the impact of fragmentation on household income. The study has shown that there is an endogenous problem and is statistically significant when estimating the relationship between the fragmentation of rice production land and household income. Therefore, applying the instrumental variable method is appropriate and reliable. The estimated results show a negative impact of fragmentation on the income of rice-producing households. This also means that increased fragmentation will reduce the income of rice-producing households in the study area. This is a huge challenge for the goal of increasing income for farmers because Land accumulation is the basis for creating a large number of agricultural products with high value, which can well meet the market demand through the application of scientific and technological advances.
From the research results, we recommend further promoting the policy "consolidation and exchange of plots" in agriculture. This is the basis for land accumulation, investment in mechanization, encouraging households with little or no labor, no conditions to invest in agricultural production development for other households to lease or transfer the right to use land for households in need. Thus, agricultural production in general and rice cultivation in particular can be developed on a large-scale commodity, and at the same time create favorable conditions to put mechanized machinery into production, improve labor productivity in agriculture. Although the study results contribute to the literature review on the impact of land fragmentation on household income, this is a case study with a small sample size and therefore requires more in-depth studies to clarify this research issue. Besides, land fragmentation is a historical issue, persists for a long time, and is associated with changes in the state's land management policies. Therefore, the use of table data is needed to clarify the causal relationship between land fragmentation and household income. Doing that in subsequent studies will make an important contribution in proposing policy directions for the development of the agricultural land market in our country.
REFERENCES
Anh, P., & Thanh, T. (2014). Initial successes of rice production in the model field model at Thuy Phu Agricultural Cooperative - Huong Thuy Town. Last accessed March 15, 2023, from <http://khuyennonghue.org.vn/guong-san-suat-gioi/tid/Nhung-thanh-cong-buo... in-mo-hinh-canh-dong-mau-tai-HTX-Nong-nghiep-Thuy-Phu--Thi-xa-Huong-Thuy/newsid/FA87463C-88A5-4BA2-921C-2987B35842DE/cid/BE42387D-85E2 -4F3E-AC7C-1DC3144EDBB0>
Blarel, B., Hazell, P., Place, F. and Quiggin, J. (1992). The economics of farm fragmentation – evidence from Ghana and Rwanda. World Bank Economic Review, 6, 233–254.
Chu, L., Nguyen, H.-T.-M., Kompas, T., Dang, K., & Bui, T. (2021). Rice land protection in a transitional economy: The case of Vietnam. Hellion, 7(4), e06754. https://doi.org/10.1016/j.heliyon.2021.e06754.
Ciaian , P. , Guri , F. , Rajcaniova , M. , Drabik , D. , & Dove , S. G. y. (2018). Land fragmentation and production diversification: A case study from rural Albania.Land Use Policy, 76, 589–599. https://doi.org/10.1016/j.landusepol.2018.02.039
Cuc, N. (2017). Land accumulation and concentration to improve competitiveness, towards a modern agriculture. Last accessed March 15, 2023, from <https://mof.gov.vn/webcenter/portal/vclvcstc/pages_r/l/chi-tiet-tin?dDoc...
Kawasaki, K. (2010). The costs and benefits of land fragmentation of rice farms in Japan*. Australian Journal of Agricultural and Resource Economics, 54(4), 509–526. https://doi.org/10.1111/j.1467-8489.2010.00509.x
Kawasaki, T., Tagawa, H., Watanabe, T., & Hanaoka, S. (2020). The effects of consolidation and privatization of ports in proximity: A case study of the Kobe and Osaka ports. Asian Journal of Shipping and Logistics, 36(1), 1–12. https://doi.org/10.1016/j.ajsl.2019.08.002
Kilic, T., Carletto, C., Miluka, J., & Savastano, S. (2009). Rural nonfarm income and its impact on agriculture: evidence from Albania. Agricultural Economics, 40(2), 139–160. https://doi.org/10.1111/j.1574-0862.2009.00366.x
Knippenberg, E., Jolliffe, D., & Hoddinott, J. (2018). Land Fragmentation and Food Insecurity in Ethiopia. Land Fragmentation and Food Insecurity in Ethiopia. World Bank, Washington, DC. https://doi.org/10.1596/1813-9450-8559
Markussen, Thomas, Finn Tarp, Do Huy Thiep, and Nguyen Do Anh Tuan. Inter- And Intra-Farmland Fragmentation In Vietnam, WIDER Working Paper 2016/11 Helsinki: UNU-WIDER, 2016.
Marsh, S. P., MacAulay, T. G., & Hung, P. V. (2007). Agricultural development and land policy in Vietnam. Last accessed 15 March 2023, from <https://www.aciar.gov.au/sites/default/files/legacy/node/770/MN123a.pdf>
Niroula, G. S., & Thapa, G. B. (2007). Impacts of land fragmentation on input use, crop yield and production efficiency in the mountains of Nepal. Land Degradation and Development (Vol. 18, Issue 3, pp. 237–248). https://doi.org/10.1002/ldr.771
Ntihinyuwa, P. D., de Vries, W. T., Chigbu, U. E., & Dukwiyiphuwe, P. A. (2019). The positive impacts of farmland fragmentation in Rwanda.Land Use Policy, 81, 565–581. https://doi.org/10.1016/j.landusepol.2018.11.005
Ntihinyurwa, P. D., & Vries, W. T. de. (2021). Farmland fragmentation concourse: Analysis of scenarios and research gaps. Land Use Policy Journal, 100. https://doi.org/10.1016/j.landusepol.2020.104936
Phan, N. T., Lee, J., & Kien, N. D. (2022). The Impact of Land Fragmentation in Rice Production on Household Food Insecurity in Vietnam. Sustainability, 14(18), 11162. https://doi.org/10.3390/su141811162
Son, N. M. (2018). Completing the system of policies and laws to promote the accumulation and concentration of land for agricultural development. Last accessed March 10, 2023, from <http://lapphap.vn/Pages/tintuc/tinchitiet.aspx?tintucid=206894>
Thu, N. T. K. (2021). Solutions to increase labor productivity in the agricultural sector in international integration. Last accessed March 15, 2023, from <https://doanhnghiepkhitesinh.vn/giai-phap-tang-gia-tri-nganh-nong-nghiep...
Tran, T. Q., & Vu, H. (2021). The impact of land fragmentation on food security in the North Central Coast, Vietnam. Asia & the Pacific Policy Studies, app5.330. https://doi.org/10.1002/app5.330
Tran, T. Q., & Vu, H. Van. (2019a). Land fragmentation and household income: First evidence from rural Vietnam. Land Use Policy, 89, 104247. https://doi.org/10.1016/j.landusepol.2019.104247
Tran, T. Q., & Vu, H. Van. (2019b). Land fragmentation and household income: First evidence from rural Vietnam. Land Use Policy, 89, 104247. https://doi.org/10.1016/j.landusepol.2019.104247
Van Hung, P., MacAulay, T. G., & Marsh, S. P. (2007). The economics of land fragmentation in the north of Vietnam. The Australian Journal of Agricultural and Resource Economics, 51(2), 195–211. https://doi.org/10.1111/j.1467-8489.2007.00378.x
Land Fragmentation Impact on Household’s Rice Income: A Case of Thua Thien Hue Province, Vietnam
ABSTRACT
The objective of the study is to analyze the influence of land fragmentation on the income of rice-producing households. The study collected information from 241 rice-producing households in two communes including Thuy Thanh and Thuy Phu in Huong Thuy town, Thua Thien Hue province. The study used Simpson's diversification index to measure land fragmentation in the study area. We applied the instrumental variable method (2-stage regression - 2sls) to overcome the endogeneity problem that may exist in the model. Research results show that the increase in land fragmentation for rice production has the effect of reducing the income of producing households. From the research results, we propose to strengthen the policy “consolidation and exchange of plots'' as a basis for land accumulation, investment in mechanization to create favorable conditions for households to produce rice on a large scale and increase the application of scientific and technical advances to production to increase income.
Keywords: Land fragmentation, rice income, instrument variable method, Thua Thien Hue
INTRODUCTION
Land is an important resource for socio-economic development activities, especially agricultural production (Phan et al., 2022). Like many countries in the world, the long-term development of Vietnam's agriculture depends heavily on the efficient use of land resources (Marsh et al., 2007). It can be seen that the current land productivity in our country is still quite low compared to other countries in the region and the world (Thu, 2021). One of the causes of that problem is the impact of land fragmentation on agricultural production (Cuc, 2017). Land fragmentation is still a prominent issue in many developing countries, including Vietnam. Rapid rural population growth and policy for equitable land access of farmer households makes the average agricultural land size of households getting smaller and smaller. Currently, our country is one of the countries with a relatively high level of land fragmentation. Specifically, Vietnam has 14.5 million farmers with nearly 70 million plots of fields, an average of 300 - 499 m2, the average household has 7 – 10 pieces, and varies greatly between regions (Son, 2018).
Huong Thuy is a town with a tradition of rice farming, located in the south of Hue city, Thua Thien Hue province. In recent years, the agricultural production policy following the model of "Small farm, Large field" (Anh, P., & Thanh, T. (2014) has received a lot of attention from the people as well as the management agencies. This policy is promoting different small farms having parcels in the same field to use the same variety and to practice the same technical protocol in order to access to mechanical services. Due to the natural, historical, and socio-economic conditions of the locality, the rice land is still fragmented and dispersed, so it is difficult to form concentrated rice production areas to apply advanced technics for high yield. This situation has affected the production efficiency of farmers because it is difficult for farmers to effectively apply mechanization and advanced agricultural technologies. Currently, some localities in Huong Thuy town are encouraging farmers to clear their fields, creating favorable conditions to apply mechanization in production, reduce labor costs, and limit pests and diseases in the production process. (Anh & Thanh, 2014). Such policies are expected to promote land accumulation and mass production of goods and create greater added value for farmers.
The literature review shows that there is a link between the current state of land fragmentation and the outputs obtained from production activities and the livelihoods of households. Land fragmentation can still have positive effects for households, for example by reducing vulnerability to the impact of climate change (Ntihinyurwa & associates, 2019). In addition, the study of Blarel and associates (Blarel, B., Hazell, P., Place, F. & Quiggin, 1992) concluded that reducing fragmentation could actually make farmers worse off. Similarly, research by Kawasaki (K. Kawasaki, 2010) shows that fragmentation reduces production risk for households. However, many studies have also shown the negative aspect of mass fragmentation to the welfare of producing households. Fragmentation can prevent households from applying new technologies in the production process (Tran and Vu, 2019a), requiring more production labor that will increase the cost of production (Ciaian et al., 2018). Some studies also show that fragmentation will make households unable to ensure food security for them (Tran & Vu, 2021; Hung et al., 2007).
As such, land fragmentation can have either a positive or a negative impact on household income, depending on the particular circumstances under consideration (Tran & Vu, 2019a). Therefore, empirical studies are needed to examine the effects of fragmentation on household income in each specific location. Therefore, it is necessary to conduct a study on the effects of rice land fragmentation on household income to provide sufficient information on the positive or negative effects of land fragmentation to the income of rice-producing households in Huong Thuy town. Research results can be the basis for developing local agricultural land management policies. In addition, the results of the study will also contribute to the theoretical and practical overview of the relationship between land fragmentation for rice production and household income.
Literature review
Many previous studies have shown that land fragmentation is a constraint to agricultural activities because fragmentation is a barrier to the use of modern, mechanized equipment, such as tractors, tillers, and harvesters (Tran & Vu, 2019a). In addition, fragmented land may impede the use of more profitable crops on a larger scale (Markussen et al., n.d.). Land fragmentation often requires more labor input for the operation of mechanized equipment, especially the considerable time spent moving between plots (Ciaian et al., 2018). Therefore, fragmented land has a significant negative impact on production efficiency and agricultural growth in many regions such as South Asia, Southeast Asia, etc. (Niroula & Thapa, 2007), (T. Kawasaki & associates, 2020). Research by Quang and Vu (Tran & Vu, 2021) found that Land fragmentation, as measured by the Simpson index, has been found to increase the likelihood that households will be food insecure during the year and have more days without protein-rich foods within a week. Such negative impacts only occur for ethnic minority households, not Kinh/Hoa households. Besides, the study of Hung et al. (Van Hung& et al., 2007) argues that land fragmentation has a negative impact on crop productivity, increased use of family labor and other monetary costs.
Although the effect of land fragmentation has been found to be negative in many studies, several other studies have shown benefits associated with land fragmentation for farm households in some cases (Ntihinyurwa & Vries, 2021). Scattered farms may be more fertile and less susceptible to disease or natural disasters (Markussen et al., 2011). Land fragmentation may reflect farmers holding multiple plots of land of varying quality, allowing them to diversify crops, allocate labor demand, and reduce production and price risks (Ciaian et al., 2018; Ntihinyurwa & Vries, 2021). A study by Ciaian et al., (2018) shows that fragmented land in rural Albania has significantly promoted agricultural diversification, thereby improving food security, and its influences on subsistence farmers are larger than on market-oriented households. Similar results are found for farm households in Rwanda, where land fragmentation has a positive impact on food quality, food sustainability and food security (Ntihinyurwa & Vries, 2021).
The literature review shows that land fragmentation can either benefit or inhibit agricultural production (Markussen et al., 2011), (Ciaian et al., 2018); but there is little evidence of an effect on household income. Land fragmentation can have a positive or negative effect on household income, depending on whether it creates a positive or negative impact on farming efficiency. On the other hand, in certain cases, the disadvantages or costs of land fragmentation may persuade farmers to diversify their livelihoods towards non-agriculture, which can yield higher returns compared to farming (Ntihinyurwa & Vries, 2021). As such, land fragmentation can have a positive or negative impact on household income, depending on the particular case considered. Therefore, a study to assess the impact of land fragmentation of rice-producing households on household income is needed.
To estimate the effect of agricultural land fragmentation on the welfare of households, many previous studies have used different models. Quang and Vu's study (Tran & Vu, 2021) used a logit model to estimate the influence of land fragmentation and other factors on food security of producing households. This can be done efficiently when the dependent variable (household food security) is a binary variable. Similarly, research rescue by Phan et al. also used an ordered probit model to estimate the influence of land fragmentation and other factors on the level of food security of producer households (Phan et al., 2022).
RESEARCH METHODS
The method of data collection
Secondary data collection: Data and information are collected from reports of the town's People's Committee over the years, the town's statistical yearbook, from various departments and local authorities, in addition, secondary data is collected through library, internet, television, and websites.
Primary data collection: The study conducted to collect data on rice growing households in 2 localities including Thuy Thanh and Thuy Phu communes. The two areas represent localities with a large total area in Huong Thuy town. The study conducted a survey of 121 farming households in Thuy Thanh commune and 120 households in Thuy commune based on a convenience sampling method.
Methods of data analysis
The study uses Simpson index to estimate land fragmentation of rice producing households in the study area: Simpson index =
. In which, the Simpson index is estimated based on the number of cells (n), the size of the cells (a) with the unit m2 and farm size (A) in m2. The Simpson index ranges from 0 to 1, with a larger index referring to more fragmentation.
To analyze the effect of fragmentation on household productivity, the study uses two methods, including linear regression (OLS) and instrumental variable method - 2-stage regression (2sls) to estimate the impact of fragmentation on earnings from manufacturing operations. The proposed theoretical model is as follows:
The objective of the study was to determine the effect of land fragmentation on the income of rice-producing households. In estimating this relationship, the problem of endogeneity between variables may arise because land fragmentation is an explanatory variable but is determined together with household income from rice production along with other factors such as production characteristics of the household. Therefore, the linear regression method can yield skewed and inconsistent results. The instrumental variable method is recommended to be used instead to give more reliable results (Tran and Vu, 2019b). This study estimates and compares the results of both the OLS method and the instrumental variable method.
To use the instrumental variable method estimation, the study uses the instrumental variable as a dummy variable such as the location of the plots (
) (Tran & Vu, 2019a) (D=1: household plots in Thuy Thanh commune, 0: household plots in Thuy Phu commune). The instrumental variable method includes two steps in the regression process. In the first step, the method estimates the association of factors belonging to households and the land fragmentation index of households. The method then estimates the effect of land fragmentation on household income from rice production. Regression model phase (2sls) are as follows:
Same as above:
is the fragmentation coefficient of rice-producing land – representing the endogenous variable, Zi are control variables such as household characteristics, ε1i, ε2i is the error Equation (2) and Equation (3) respectively. The conditions of IV are expressed through the following system of equations:
Thus, the instrumental variable D must satisfy two conditions: (i) D is not correlated with
; (ii) but D is also correlated with the fragmentation coefficient Simpson. To determine whether the problem of endogeneity exists or not, the study used Wu-Hausman statistical test with the hypothesis H0: The model does not exist endogenous phenomenon, and H1: The model exists endogenous phenomenon. If the P-value is less than 5%, this means that there is an endogeneity problem in linear regression (OLS). In addition, the Cragg-Donald Wald test is used to evaluate the strength of the selected instrument variable.
The study also used non-parametric analysis to graph the distribution of income variables in relation to the change of Simpson's land fragmentation coefficient, in five percentiles (0; 0.25; 0.5; 0.75; 1). Non-parametric testing methods can be used quickly, easily, and just as effectively as non-parametric testing methods for data for which the distribution function has not been determined. This method will give an initial view of the relationship between the variables of interest, thereby laying the foundation for more in-depth analytical methods.
RESULTS AND DISCUSSION
Using agricultural land situation in Huong Thuy town, Thua thien Hue province
In Huong Thuy town, the total land area used for agricultural production is 35,464 ha, accounting for 78% of the local natural land area (Table 1). In which, the total area of rice cultivation is 3,064 ha, accounting for 6.74% of the total land area in Huong Thuy town. Besides, the area for other agricultural production accounted for a fairly large proportion with 70.60%. In general, Huong Thuy town has a diversity of forms of land use in agricultural production.
Situation of rice production in Huong Thuy town
As shown in Table 2, in Huong Thuy town, in 2019, the Rice cultivated area of the whole town was 6,154 ha, not much different from the 2018 area of 6,215 ha; but rice production in 2019 is 39,321 tons, an increase of 430 tons compared to 2018. Winter-spring crop in 2018 - 2019, the cultivated area of the whole town is 3,124 hectares, with varieties like HT1, TBR279, of which, the area of rice production accounts for more than 10%. Although the weather at the beginning of the crop was quite unfavorable, heavy rain lasted for many days, but thanks to the initiative of the agricultural sector and the attention of leaders at all levels, farmers not only did not lose their crops but also achieved good results of a bountiful crop, 85% of Huong Thuy's 2019 winter-spring rice area is well developed, the field uniformity is quite high, and the crop is bountiful. In 2018, the total rice production area was 6,215 ha, of which winter-spring rice accounted for 3,124 ha, summer-autumn rice accounted for 3,091 ha. In 2019 compared to 2018 it decreased by 61 hectares, equivalent to 99.03%. Winter-spring rice and summer-autumn rice decreased by 32 hectares and 29 hectares, respectively, accounting for 98.98% and 99.08%. In terms of output, in 2019 compared to 2018, the total output increased by 430 tons, equivalent to 101.10%. Winter-spring rice decreased by 287 tons compared to 2018 of 20,379 tons. Summer-autumn rice fluctuated sharply to 716 tons, equivalent to 103.87%. This figure shows that in the summer-autumn crop in 2019 compared to 2018 there is an improvement in seed quality as well as favorable natural conditions for cultivation. In 2020, the area of all crops decreased due to the influence of many factors affecting the farming process, the total area decreased by 59 hectares compared to 2019 to 6,154 hectares, equivalent to 99.20%. In which, the winter-spring crop decreased by 35 hectares, accounting for 98.86%; Summer-autumn crop reduced 24 hectares accounted for 99.55%. Output in this year also decreased significantly because of the prolonged disease situation, affecting many aspects of farmers, especially the situation of rice trading after harvest, down 1,822 tons for the whole year and decreased by 1,996 tons in the summer-autumn crop. Productivity in 2019 compared to 2018 increased by 1.32 quintals/ha. It's a small number, but it affects the whole production process. But due to the COVID-19 pandemic, the yield in 2020 compared to 2019 decreased by 2.4 quintals/ha.
Basic information about surveyed households
A survey of 241 rice-producing households in two areas shows that the fragmentation coefficient in Thuy Phu commune is lower than that in Thuy Thanh commune, reaching 0.39 and 0.42, respectively (Table 3). The average age of the household head is over 50 years old. It is a traditional field and requires experienced people to produce, so the age of the household head will be quite old. However, the education level of the household head is not high, about 7 to 9 years on average. In the countryside, the main occupation is agriculture, the issue of cultural level is often not paid due attention by the farmer households. What farmers care about is production skills and experience. Besides, the data on the labor force in rice production is about 2 members per household and the number of man-days in rice production is about 71 man-days per household. The average yield of rice-producing households is 6.65 tons/ha. Therefore, there was a similarity in productivity between surveyed households in the 2 study areas (Table 3).
The survey also shows that the majority of household heads in rice production are predominantly male, with over 70% of household heads interviewed being male. A large percentage of households have adopted new rice varieties to mitigate some of the risks in production due to the effects of climate change. This also helps households reduce investment costs to reduce risks in production. The percentage of households using new rice varieties is 67.5% of the total number of households interviewed. Credit also plays an important role in agricultural production. Access to credit helps households have enough financial resources to pay for input production costs. The percentage of households accessing credit accounts for over 90% of the total number of households interviewed.
Impact of land fragmentation on household income
Non-parametric analysis results
The results of the non-parametric analysis are presented in Figure 1, in which the household income from rice cultivation is presented according to the five percentiles of the Simpson coefficient of fragmentation: 0-0.25;0.25-0.5;0.5-0.75 and 0.75-1. The results show that at the low percentile level, corresponding to a low fragmentation coefficient (0-0.25), the income of households from rice cultivation tends to be the highest. This income distribution tends to decrease gradually with higher percentiles such as 0.25-0.5; 0.5-0.75; and 0.75-1. These initial results show that there exists a strong relationship between income and land fragmentation. However, to clarify and find statistical evidence of this relationship, applying more complex statistical methods is necessary and we present in the next step.
Regression Result OLS and 2SLS
The study estimated the impact of productive land fragmentation on household income by two methods: linear regression (OLS) and instrumental variable method. The p-value of the Wu-Hausman test for endogeneity is less than 5%. Therefore, there is an endogenous problem when estimating the impact of land fragmentation on household income. Therefore, estimation by instrumental variable method is necessary. The Cragg-Donald Wald (F) statistic is greater than 10, so the instrumental variable method results are likely to be reliable.
Estimation results of the regression model by means of instrumental variables show that increased land fragmentation has a negative impact on household income and is statistically significant with a coefficient of -2.994 at the level of statistical significance 5% (Table 4). This result is similar to the study of Tran and Vu (Tran & Vu, 2019b). This is explained as when land is divided into smaller pieces, it can be difficult to efficiently use resources such as labor, machinery and capital for production. In addition, fragmented land can lead to difficulties in managing and maintaining efficiency in the production process, thereby reducing income generation potential for households. Land fragmentation is also a factor preventing people and businesses from making long-term, large investments in agriculture.
The reality is that farmers cannot determine the market price of their output, so increasing production can be a way to help producers gain more profits. In addition, if farmers can produce rice on a large scale with low cost, they can increase profit margin, which will also contribute to increasing household income. With the strong development of scientific and technical progress, farmers can increase rice production through the use of new varieties, new techniques, etc. These measures are only really effective when households farming on a large scale to increase the efficiency of input use in the production process. The accumulation of land has contributed to reducing the small-scale production area, creating a large number of agricultural products, high income value and creating favorable conditions for farmers to participate in production and application, use science and technology. Reducing land fragmentation in rice production is an important strategy to develop Vietnam's rice production in the direction of large commodities.
CONCLUSIONS AND RECOMMENDATIONS
The objective of the study is to analyze the effect of land fragmentation on the income of rice production. A survey of 241 rice-producing households in two areas shows that the fragmentation coefficient in Thuy Phu commune is lower than that in Thuy Thanh commune, reaching 0.39 and 0.42, respectively. Two methods are used in the study, including: linear regression (OLS) and instrumental variable method to analyze the impact of fragmentation on household income. The study has shown that there is an endogenous problem and is statistically significant when estimating the relationship between the fragmentation of rice production land and household income. Therefore, applying the instrumental variable method is appropriate and reliable. The estimated results show a negative impact of fragmentation on the income of rice-producing households. This also means that increased fragmentation will reduce the income of rice-producing households in the study area. This is a huge challenge for the goal of increasing income for farmers because Land accumulation is the basis for creating a large number of agricultural products with high value, which can well meet the market demand through the application of scientific and technological advances.
From the research results, we recommend further promoting the policy "consolidation and exchange of plots" in agriculture. This is the basis for land accumulation, investment in mechanization, encouraging households with little or no labor, no conditions to invest in agricultural production development for other households to lease or transfer the right to use land for households in need. Thus, agricultural production in general and rice cultivation in particular can be developed on a large-scale commodity, and at the same time create favorable conditions to put mechanized machinery into production, improve labor productivity in agriculture. Although the study results contribute to the literature review on the impact of land fragmentation on household income, this is a case study with a small sample size and therefore requires more in-depth studies to clarify this research issue. Besides, land fragmentation is a historical issue, persists for a long time, and is associated with changes in the state's land management policies. Therefore, the use of table data is needed to clarify the causal relationship between land fragmentation and household income. Doing that in subsequent studies will make an important contribution in proposing policy directions for the development of the agricultural land market in our country.
REFERENCES
Anh, P., & Thanh, T. (2014). Initial successes of rice production in the model field model at Thuy Phu Agricultural Cooperative - Huong Thuy Town. Last accessed March 15, 2023, from <http://khuyennonghue.org.vn/guong-san-suat-gioi/tid/Nhung-thanh-cong-buo... in-mo-hinh-canh-dong-mau-tai-HTX-Nong-nghiep-Thuy-Phu--Thi-xa-Huong-Thuy/newsid/FA87463C-88A5-4BA2-921C-2987B35842DE/cid/BE42387D-85E2 -4F3E-AC7C-1DC3144EDBB0>
Blarel, B., Hazell, P., Place, F. and Quiggin, J. (1992). The economics of farm fragmentation – evidence from Ghana and Rwanda. World Bank Economic Review, 6, 233–254.
Chu, L., Nguyen, H.-T.-M., Kompas, T., Dang, K., & Bui, T. (2021). Rice land protection in a transitional economy: The case of Vietnam. Hellion, 7(4), e06754. https://doi.org/10.1016/j.heliyon.2021.e06754.
Ciaian , P. , Guri , F. , Rajcaniova , M. , Drabik , D. , & Dove , S. G. y. (2018). Land fragmentation and production diversification: A case study from rural Albania.Land Use Policy, 76, 589–599. https://doi.org/10.1016/j.landusepol.2018.02.039
Cuc, N. (2017). Land accumulation and concentration to improve competitiveness, towards a modern agriculture. Last accessed March 15, 2023, from <https://mof.gov.vn/webcenter/portal/vclvcstc/pages_r/l/chi-tiet-tin?dDoc...
Kawasaki, K. (2010). The costs and benefits of land fragmentation of rice farms in Japan*. Australian Journal of Agricultural and Resource Economics, 54(4), 509–526. https://doi.org/10.1111/j.1467-8489.2010.00509.x
Kawasaki, T., Tagawa, H., Watanabe, T., & Hanaoka, S. (2020). The effects of consolidation and privatization of ports in proximity: A case study of the Kobe and Osaka ports. Asian Journal of Shipping and Logistics, 36(1), 1–12. https://doi.org/10.1016/j.ajsl.2019.08.002
Kilic, T., Carletto, C., Miluka, J., & Savastano, S. (2009). Rural nonfarm income and its impact on agriculture: evidence from Albania. Agricultural Economics, 40(2), 139–160. https://doi.org/10.1111/j.1574-0862.2009.00366.x
Knippenberg, E., Jolliffe, D., & Hoddinott, J. (2018). Land Fragmentation and Food Insecurity in Ethiopia. Land Fragmentation and Food Insecurity in Ethiopia. World Bank, Washington, DC. https://doi.org/10.1596/1813-9450-8559
Markussen, Thomas, Finn Tarp, Do Huy Thiep, and Nguyen Do Anh Tuan. Inter- And Intra-Farmland Fragmentation In Vietnam, WIDER Working Paper 2016/11 Helsinki: UNU-WIDER, 2016.
Marsh, S. P., MacAulay, T. G., & Hung, P. V. (2007). Agricultural development and land policy in Vietnam. Last accessed 15 March 2023, from <https://www.aciar.gov.au/sites/default/files/legacy/node/770/MN123a.pdf>
Niroula, G. S., & Thapa, G. B. (2007). Impacts of land fragmentation on input use, crop yield and production efficiency in the mountains of Nepal. Land Degradation and Development (Vol. 18, Issue 3, pp. 237–248). https://doi.org/10.1002/ldr.771
Ntihinyuwa, P. D., de Vries, W. T., Chigbu, U. E., & Dukwiyiphuwe, P. A. (2019). The positive impacts of farmland fragmentation in Rwanda.Land Use Policy, 81, 565–581. https://doi.org/10.1016/j.landusepol.2018.11.005
Ntihinyurwa, P. D., & Vries, W. T. de. (2021). Farmland fragmentation concourse: Analysis of scenarios and research gaps. Land Use Policy Journal, 100. https://doi.org/10.1016/j.landusepol.2020.104936
Phan, N. T., Lee, J., & Kien, N. D. (2022). The Impact of Land Fragmentation in Rice Production on Household Food Insecurity in Vietnam. Sustainability, 14(18), 11162. https://doi.org/10.3390/su141811162
Son, N. M. (2018). Completing the system of policies and laws to promote the accumulation and concentration of land for agricultural development. Last accessed March 10, 2023, from <http://lapphap.vn/Pages/tintuc/tinchitiet.aspx?tintucid=206894>
Thu, N. T. K. (2021). Solutions to increase labor productivity in the agricultural sector in international integration. Last accessed March 15, 2023, from <https://doanhnghiepkhitesinh.vn/giai-phap-tang-gia-tri-nganh-nong-nghiep...
Tran, T. Q., & Vu, H. (2021). The impact of land fragmentation on food security in the North Central Coast, Vietnam. Asia & the Pacific Policy Studies, app5.330. https://doi.org/10.1002/app5.330
Tran, T. Q., & Vu, H. Van. (2019a). Land fragmentation and household income: First evidence from rural Vietnam. Land Use Policy, 89, 104247. https://doi.org/10.1016/j.landusepol.2019.104247
Tran, T. Q., & Vu, H. Van. (2019b). Land fragmentation and household income: First evidence from rural Vietnam. Land Use Policy, 89, 104247. https://doi.org/10.1016/j.landusepol.2019.104247
Van Hung, P., MacAulay, T. G., & Marsh, S. P. (2007). The economics of land fragmentation in the north of Vietnam. The Australian Journal of Agricultural and Resource Economics, 51(2), 195–211. https://doi.org/10.1111/j.1467-8489.2007.00378.x