The Impact of Natural Disasters and Pest Infestations on Technical Efficiency in Rice Production: A Study in Vietnam

Abstract

This study examines the effects of natural disasters, such as typhoons, floods, droughts, and pest infestations, on the technical efficiency of rice production in Vietnam. Employing stochastic frontier analysis (SFA), the research estimates the technical efficiency in rice production of 2394 farmers from the 2018 Vietnam Access to Resources Household Survey (VARHS) dataset. The findings indicate that the average technical efficiency of rice production among these farmers is 78.99%. Exposure to natural disasters and pest infestations leads farmers to reduce their investments in rice production, resulting in decreased technical efficiency, lower yields, and reduced profitability. Among the various disasters, droughts have the most significant adverse impact on technical efficiency in rice production. The results highlight the limited capacity of farmers to cope with the challenges posed by natural disasters in rice production. The study emphasizes the importance of providing timely support to farmers, fostering resilience within the context of rice farming, and enhancing agricultural sustainability in Vietnam. To address these challenges effectively, policymakers are advised to prioritize facilitating farmers’ access to agricultural insurance. Additionally, encouraging income diversification among farmers becomes crucial to ensuring provisions in the case of income loss from rice production due to natural disasters or pest infestations. Moreover, measures such as promoting climate-smart agricultural practices, improving water management infrastructure, establishing early warning systems, and emphasizing pest and disease control measures can be implemented to mitigate losses resulting from natural disasters and pest infestations.

1. Introduction

Vietnam’s economy heavily relies on agriculture, with this sector employing 34.5% of the workforce and contributing 13.96% to the country’s gross domestic product. Rice cultivation plays a pivotal role, covering 7.47 million hectares and producing 43.5 million tons in 2019 [1]. Vietnam, positioned as the world’s third-largest rice exporter, assumes a vital role in safeguarding global food security. Nonetheless, the sustainability of rice farming in Vietnam encounters significant challenges, particularly regarding the income instability of local farmers. The volatile nature of their earnings stems from multiple production risk factors, including the adverse impacts of natural disasters such as typhoons, floods, and droughts, along with pest infestations and unpredictable output prices [2]. These risks, predominantly beyond the farmers’ control, bear implications not only for their livelihoods but also for the long-term viability and sustainability of the rice sector in Vietnam [3].

Rice production, as a natural resource-dependent and climate-dependent industry, is susceptible to the worsening risks caused by climate change [4,5,6]. Climate change and global warming contribute to the heightened frequency of natural disasters worldwide [7]. Vietnam, located in the tropical cyclone belt, has experienced significant impacts on rice production due to a series of natural disasters [8]. Typhoons, floods, droughts, sea level rise, and saline intrusion are among the various natural disasters that affect rice production in Vietnam [9]. Moreover, global warming has intensified water circulation, leading to more frequent and severe extreme weather events such as floods and droughts in this country [10].

Additionally, Vietnam’s long and narrow terrain, as well as its extensive coastline, increase its vulnerability to natural disasters and climate change. The country encounters over 10 typhoons each year, resulting in substantial damage to agricultural output through floods and landslides. In 2017, the estimated cost of natural disasters amounted to approximately EUR 2.47 billion, with 246.2 thousand hectares of rice fields being flooded and adversely affected. In 2018, the area of damaged rice cultivation increased to 261.1 thousand hectares [11]. Furthermore, drought and saltwater intrusion pose detrimental consequences for rice production, as evident by the fact that 34,000 hectares of rice in the Mekong River Delta suffered from such effects in 2019 [1].

Studies project a decline in Vietnam’s rice production of up to 18% by 2030 without interventions [12]. Production risks associated with natural disasters and climate change are estimated to reduce rice farmers’ income by 16.02% [4]. However, implementing climate-smart agriculture practices can mitigate these negative impacts and enhance technical efficiency. Participating in climate-smart agriculture programs has been found to improve technical efficiency by 5% to 8% compared to non-participating farmers [4].

In addition to natural disasters, Vietnamese rice farming faces a variety of pest and disease hazards. The tropical climate of Vietnam creates favorable conditions for the proliferation of pests and molds that can harm crops [9]. Moreover, excessive use of urea fertilizer attracts harmful insects and induces diseases in rice plants, leading to increased pesticide application and higher production costs for farmers [13]. Empirical studies indicate that Vietnamese rice farmers typically apply pesticides 6.1 times per crop, with detrimental effects on both the environment and farmers’ health [14]. Notably, pest infestations such as brown planthoppers and yellow dwarf disease, the latter of which has no known cure, seriously jeopardize rice production in Vietnam. The 2017–2018 outbreak of yellow dwarf leaf on rice, which resurfaced in southern regions of Vietnam, resulted in crop damage and the infection of thousands of hectares of rice fields [11].

Given these multiple challenges, assessing the impacts of natural disasters and pest infestations on rice farming efficiency becomes imperative. Existing literature suggests that natural disasters and pest infestations reduce output and increase the use of inputs in rice production, thereby negatively affecting technical efficiency [15,16] Technical efficiency refers to a farmer’s capacity to attain the highest possible output utilizing existing and accessible technology [17,18]. Numerous studies have been conducted to examine the technical efficiency of rice production in Vietnam and the factors that influence it [4,19,20,21,22,23]. However, there is a limited understanding of the effects of natural disasters and pest infestations on technical efficiency in rice production in Vietnam and how farmers overcome the consequences of these challenges.

The primary objective of this study is to comprehensively evaluate the impacts of natural disasters, including typhoons, floods, and droughts, as well as pest infestations, on technical efficiency in rice production in Vietnam. Through empirical evidence, it enhances the existing literature by revealing the consequences of these events on rice production efficiency. Furthermore, the study investigates farmers’ financial losses and adaptive responses to these challenges. The findings of this study provide valuable insights and recommendations for enhancing rice production efficiency in Vietnam, aligning with the goals of sustainability and resilience in agricultural systems.

2. Materials and Methods

2.1. Stochastic Frontier Analysis (SFA)

To assess the technical efficiency of rice production and analyze the impact of natural disasters and pest infestations, this study employed the stochastic frontier analysis (SFA) method. The SFA method, introduced by [24,25], incorporates a production function with 2 error term components: random effect and technical inefficiency. The maximum likelihood estimation of the production function, considering these 2 error terms, is used to determine the technical efficiency. The model is represented as follows:

Y_i = X_iβ + (V_i − U_i)

(1)

where Y_i denotes the production output of the ith farm; X_i is a 1 × k vector of input quantities for the ith farm; β represents the parameter vector; V_i signifies the random error assumed to follow the N(0, σ_v²) distribution; and U_i represents the non-negative technical inefficiency effect following the truncation at 0 of the N(μ_i, σ_u²) distribution. The estimation of the technical inefficiency function is formulated as follows:

μ_i = z_iδ + e_i

(2)

where z_i is a p × 1 vector of variables that may influence inefficiency; δ is a 1 × p vector of parameters to be estimated; and e_i is the estimation error of the model. The technical inefficiency of each farmer can be estimated using the conditional distribution of u_i with appropriate values of ε_i and the corresponding parameters [26]. The formula for determining the technical efficiency of individual farmers is given by:

$${\mathrm{TE}}_{\mathrm{i}}=\exp (-\widehat{{\mathrm{u}}_{\mathrm{i}}})=\exp (-\mathrm{E}({\mathrm{u}}_{\mathrm{i}} | {\mathsf{\epsilon }}_{\mathrm{i}}))$$

(3)

where TE_i falls between 0 and 1. The study utilized the FRONTIER software, version 4.1c, to estimate the function parameters and the technical efficiency of individual farmers.

Based on the theoretical framework of the SFA model, this study utilized the Cobb–Douglas function, encompassing 7 independent input variables to estimate the technical efficiency of rice production in Vietnam:

lnYield_i = β₀ + β₁lnSeed_i + β₂lnChemifer_i + β₃lnOrgafer_i + β₄lnPesticide_i + β₅lnH_Labor_i + β₆lnOutLabor_i +
β₇lnMachinery_i + V_i − U_i

(4)

where lnYield_i is the logarithm of rice yield per hectare of the i^th farm; input variables are the logarithm of the number of inputs used per hectare, such as the logarithm of seed cost (lnSeed_i), chemical fertilizer cost (lnChemifer_i), organic fertilizer cost (lnOrgafer_i), pesticide cost (lnPesticide_i), household labor (lnH_Labor_i), outsourced labor cost (lnOutLabor_i), and hiring machinery cost (lnMachinery_i). The measurement of technical efficiency focuses on assessing the quantity of inputs and outputs without considering their respective prices [17]. However, the VARHS dataset has limitations, as it does not provide specific quantities (in kilograms per hectare) for each input. Therefore, this model uses the cost of inputs as a substitute for input quantities.

The technical inefficiency model incorporates 3 groups of variables related to natural disasters and pest infestations, farmer characteristics, and regional factors, as outlined below:

$$\begin{gathered} {\mathsf{\mu }}_{\mathrm{i}}={\mathsf{\delta }}_0+{\mathsf{\delta }}_1{\mathrm{Pest}}_{\mathrm{i}}+{\mathsf{\delta }}_2{\mathrm{Flood}}_{\mathrm{i}}+{\mathsf{\delta }}_3{\mathrm{Drought}}_{\mathrm{i}}+{\mathsf{\delta }}_4{\mathrm{Typhoon}}_{\mathrm{i}}+{\mathsf{\delta }}_5{\mathrm{Ethnicity}}_{\mathrm{i}}+{\mathsf{\delta }}_6{\mathrm{Gender}}_{\mathrm{i}}+{\mathsf{\delta }}_7{\mathrm{Age}}_{\mathrm{i}}+{\mathsf{\delta }}_8{\mathrm{Education}}_{\mathrm{i}}+ \\ {{\displaystyle \sum }}_{\mathrm{n}=1}^{11}{\mathsf{\alpha }}_{\mathrm{n}}{\mathrm{Province}}_{\mathrm{ni}}+{\mathrm{e}}_{\mathrm{i}} \end{gathered}$$

(5)

where Pest_i represents a dummy variable equal to 1 if farmers experienced pest infestations and 0 otherwise; Flood_i has a value of 1 if floods had an impact on the farmer’s rice production and 0 otherwise; Drought_i is also a dummy variable, equal to 1 if farmers suffered from drought damage and 0 otherwise; Typhoon_i takes on a value of 1 if typhoons swept over the farmer’s rice field and 0 otherwise; Ethnicity_i is valued at 1 if the farmer is Kinh ethnicity and 0 otherwise; Gender_i is equal to 1 if the farmer is male and 0 if the farmer is female; Age_i is the number of years of age of the farmer; Education_i indicates the years of schooling of the farmer; Province_ni are dummy variables related to provinces in VARHS, in which Ha Tay province (which has been merged into Ha Noi since 2008) is used for comparison; δ_i and α_n are unknown parameters that need to be estimated; and e_i represents the estimation error of the model.

Table 1. A system of variables used in the SFA model.

Type of Variables	Variables	Unit	Explanation
Output	Yield	Kg per ha	Rice yield per hectare
Input	Seeds	EUR per ha	Seed cost per hectare
	Chemical fertilizers	EUR per ha	Chemical fertilizer cost per hectare
	Organic fertilizers	EUR per ha	Organic fertilizer cost per hectare
	Pesticides	EUR per ha	Pesticide cost per hectare
	Household labor	Days per ha	Number of household labor employed per hectare
	Outsourced labor	EUR per ha	Outsourced labor cost per hectare
	Machinery	EUR per ha	Machinery cost per hectare
Inefficiency	Pest infestations	Dummy	1 = Loss due to pest infestations, 0 = No loss due to pest infestations
	Flood	Dummy	1 = Loss due to flood, 0 = No loss due to flood
	Drought	Dummy	1 = Loss due to drought, 0 = No loss due to drought
	Typhoon	Dummy	1 = Loss due to typhoon, 0 = No loss due to typhoon
	Ethnicity	Dummy	1 = Kinh, 0 = Other ethnicity
	Gender	Dummy	1 = male, 0 = female
	Age	Year	The age of household head
	Education	Year	Number of years of schooling of the household head
	Province1	Dummy	1 = Lao Cai, 0 = Other province
	Province2	Dummy	1 = Phu Tho, 0 = Other province
	Province3	Dummy	1 = Lai Chau, 0 = Other province
	Province4	Dummy	1 = Dien Bien, 0 = Other province
	Province5	Dummy	1 = Nghe An, 0 = Other province
	Province6	Dummy	1 = Quang Nam, 0 = Other province
	Province7	Dummy	1 = Khanh Hoa, 0 = Other province
	Province8	Dummy	1 = Dak Lak, 0 = Other province
	Province9	Dummy	1 = Dak Nong, 0 = Other province
	Province10	Dummy	1 = Lam Dong, 0 = Other province
	Province11	Dummy	1 = Long An, 0 = Other province

summarizes the variables included in the SFA model.

2.2. Data Collection

The data utilized for this study were obtained from the Vietnam Access to Resources Household Survey (VARHS) conducted in 2018. The survey employed a stratified sampling approach with multiple stages to ensure a representative sample of Vietnamese households. The survey covered rural households across 12 provinces in 5 regions of Vietnam: the Red River Delta (Ha Tay), the North (Lao Cai, Phu Tho, Lai Chau, and Dien Bien), the Central Coast (Nghe An, Quang Nam, and Khanh Hoa), the Central Highlands (Dak Lak, Dak Nong, and Lam Dong), and the Mekong River Delta (Long An). Data were collected from 3807 households located in 485 communes. The survey took place from June to July 2018 and focused on gathering information regarding the households’ experiences over the past 12 months. Figure 1 illustrates the geographical distribution of the 12 provinces included in the VARHS 2018 [27].

This study aims to estimate the technical efficiency of rice production and analyze the factors influencing it. Hence, only data from rice farmers were selected for analysis. After filtering the dataset, a total of 2394 rice farmers were included in the study. Based on the extent of damage caused by natural disasters and pest infestations, these farmers were classified into 2 groups: affected and unaffected farmer groups.

Table 2. Distribution of observations by affected and unaffected farmer groups in each province.

Provinces	Observations	Unaffected Farmer Group		Affected Farmer Group
		Frequency	Percentage	Frequency	Percentage
Ha Tay	301	247	82.06	54	17.94
Lao Cai	324	197	60.80	127	39.20
Phu Tho	236	214	90.68	22	9.32
Lai Chau	382	260	68.06	122	31.94
Dien Bien	365	263	72.05	102	27.95
Nghe An	134	56	41.79	78	58.21
Quang Nam	190	186	97.89	4	2.11
Khanh Hoa	30	10	33.33	20	66.67
Dak Lak	198	122	61.62	76	38.38
Dak Nong	99	73	73.74	26	26.26
Lam Dong	13	4	30.77	9	69.23
Long An	122	103	84.43	19	15.57
Total	2394	1735	72.47	659	27.53

presents the distribution of farmers within the sample, revealing that the rice production of 1735 farmers (72.47%) was unaffected by natural disasters and pest infestations. The affected farmer group, comprising those who experienced losses due to natural disasters and pest infestations in rice production, accounted for 27.53% of the farmers surveyed.

Due to variations in the number of rice farmers across provinces, the quantity of observations collected differs among provinces. Particularly, Lai Chau province stood out with the highest number of observations, encompassing 382 households. Other provinces with a substantial number of observations included Dien Bien, Lao Cai, and Ha Tay, each with over 300 households. In contrast, the provinces of Lam Dong and Khanh Hoa exhibited the lowest number of observations, comprising only 13 and 30 rice-farming households, respectively.

The percentage of farmers experiencing losses due to natural disasters and pest infestations in rice production also varies among provinces. Notably, Nghe An, Khanh Hoa, and Lam Dong provinces stand out as regions where more than half of the observations reported production losses during rice production. Similarly, Lao Cai, Dak Lak, and Lai Chau provinces had a significant number of affected farmers, accounting for 39.2%, 38.38%, and 31.94% of the total observations in each respective province. In contrast, Quang Nam and Phu Tho provinces had a relatively lower proportion of affected farmers, constituting less than 10% of the province’s observations.

Table 3. Number of affected farmers by pest infestations, floods, droughts, and typhoons in each province.

Provinces	Pest Infestations		Floods		Droughts		Typhoons
	Frequency	Percentage	Frequency	Percentage	Frequency	Percentage	Frequency	Percentage
Ha Tay (n = 301)	34	11.30	12	3.99	2	0.66	10	3.32
Lao Cai (n = 324)	55	16.98	4	1.23	71	21.91	27	8.33
Phu Tho (n = 236)	14	5.93	4	1.69	3	1.27	3	1.27
Lai Chau (n = 382)	91	23.82	7	1.83	3	0.79	39	10.21
Dien Bien (n = 365)	87	23.84	10	2.74	5	1.37	22	6.03
Nghe An (n = 134)	42	31.34	8	5.97	18	13.43	46	34.33
Quang Nam (n = 190)	3	1.58	1	0.53	0	0.00	0	0.00
Khanh Hoa (n = 30)	0	0.00	0	0.00	0	0.00	20	66.67
Dak Lak (n = 198)	33	16.67	7	3.54	6	3.03	45	22.73
Dak Nong (n = 99)	7	7.07	2	2.02	5	5.05	15	15.15
Lam Dong (n = 13)	1	7.69	3	23.08	3	23.08	2	15.38
Long An (n = 122)	18	14.75	0	0.00	0	0.00	1	0.82
Total (n = 2394)	385	16.08	58	2.42	116	4.85	230	9.61

presents detailed statistics pertaining to the number of farmers affected by pest infestations and natural disasters, including floods, droughts, and typhoons, in each province. The data indicate that pest infestations affected 385 rice farmers, constituting 16.08% of the study sample. It is noteworthy that all farmers in the sample employed pesticides during rice cultivation. This indicates the presence of pest and disease challenges among all surveyed farmers. However, farmers who were capable of effectively utilizing pesticides to control pests in their rice fields were classified as unaffected by such pest infestations. Conversely, farmers experiencing a loss of yield and income from rice due to pests exceeding their control measures were considered affected by pest infestations. In addition, typhoons resulted in financial losses in rice production for 9.61% of farmers, while droughts caused losses for 4.85% of farmers, and floods impacted 2.42% of farmers.

The number of farmers affected by pest infestations, droughts, floods, and typhoons showed variations between provinces. Particularly, Nghe An (31.34%), Dien Bien (23.84%), and Lai Chau (23.82%) provinces had a higher likelihood of experiencing pest infestations compared to other provinces. Lam Dong province had the highest percentage (23.08%) of facing droughts during the dry season and floods during the rainy season. Furthermore, drought impacted 21.91% of farmers in Lao Cai province. Khanh Hoa province exhibited the highest proportion of typhoons, with a rate of 66.67%.

Table 4. Money losses per household due to pest infestations, floods, droughts, and typhoons.

Loss Due to Risks (EUR per Household)	Pest Infestations (n = 385)	Floods (n = 58)	Droughts (n = 116)	Typhoons (n = 230)	Average Total
Ha Tay	281.1	337.8	79.1	564.5	359.5
Lao Cai	455.1	61.1	64.1	232.1	284.2
Phu Tho	246.1	209.5	104.2	35.6	213.8
Lai Chau	95.7	271.0	36.9	67.0	109.2
Dien Bien	246.8	46.4	81.3	247.8	272.5
Nghe An	53.6	246.7	105.3	119.6	149.0
Quang Nam	307.2	79.1	-	-	250.2
Khanh Hoa	-	-	-	418.1	418.1
Dak Lak	437.6	804.1	236.9	563.5	616.5
Dak Nong	352.1	237.3	1407.4	650.6	759.0
Lam Dong	19.8	158.2	167.5	11.9	113.4
Long An	677.9	-	-	791.1	683.9
Average	261.1	278.1	141.3	303.8	307.9

presents the average monetary losses per household due to pest infestations and natural disasters. It is important to note that a household may experience multiple natural disasters and pest infestations within a year. The average total loss per household amounts to EUR 307.9. Typhoons caused the most damage, with an average loss of EUR 303.8 per household, followed by floods with EUR 278.1 per household and pest infestations with EUR 261.1 per household. Droughts incurred the lowest average loss at EUR 141.3 per household.

The extent of losses resulting from pest infestations and natural disasters varies not only between provinces but also within provinces. Long An province, for instance, displayed the highest average household losses attributed to pest infestations (EUR 677.9 per household) and typhoons (EUR 791.1 per household). However, floods and droughts did not impact rice production in Long An in 2018. Pest infestations also heavily affected rice farming in Lao Cai and Dak Lak provinces, resulting in losses of EUR 455.1 and EUR 437.6 per household, respectively. Additionally, floods caused a significant loss in household incomes in Dak Lak, with a reduction of EUR 804.1 per household.

On the other hand, drought had the most severe consequences in Dak Nong, where losses reached EUR 1407.4 per household. Notably, typhoons emerged as the primary cause of significant losses across multiple provinces. Long An, Dak Nong, Ha Tay, and Dak Lak provinces suffered losses exceeding EUR 500 per household as a result of typhoons. In Khanh Hoa province, which remained unaffected by other natural disasters and pest infestations, losses due to typhoons were estimated to be EUR 418.1 per household.

3. Results and Discussions

3.1. Summary Statistics

Table 5. Characteristics of two farmer groups.

Items	Group	Unaffected Farmer Group		Affected Farmer Group
		Frequency (n = 1735)	Percentage	Frequency (n = 659)	Percentage
Age	≤40	317	18.27	141	21.40
	41–60	1015	58.50	394	59.79
	≥61	403	23.23	124	18.82
Gender	Female	284	16.37	101	15.33
	Male	1451	83.63	558	84.67
Ethnicity	Kinh	849	48.93	225	34.14
	Others	886	51.07	434	65.86
Education	Uneducated	323	18.62	148	22.46
	Primary	483	27.84	192	29.14
	Secondary	688	39.65	249	37.78
	High school	166	9.57	52	7.89
	Intermediate	39	2.25	12	1.82
	College	11	0.63	4	0.61
	Bachelor	24	1.38	2	0.30
	Postgraduate	1	0.06	0	0.00

presents the demographic characteristics of two farmer groups. Both groups showed similarities in terms of age, gender, ethnicity, and education level. Particularly, the majority of farmers in both groups fell within the age range of 40 to 60 years. Male farmers constituted a majority in both the unaffected and affected groups, comprising 83.63% and 84.67%, respectively. The Kinh ethnicity accounted for 34.14% and 48.93% of farmers in the affected and unaffected groups, respectively. Vietnam is home to 54 ethnic groups, with the Kinh people representing 85.3% of the country’s population, while the remaining ethnic groups are considered ethnic minorities [1]. Therefore, the analysis focuses on the technical efficiency between the majority ethnic group (Kinh) and the ethnic minorities. Regarding education, the majority of farmers in both groups had achieved secondary school education or lower (90% overall). However, there were differences in the proportion of uneducated farmers. In the unaffected group, 18.62% of farmers were uneducated, while in the affected group, the percentage was slightly higher at 22.46%.

Table 6. Rice production characteristics of the two farmer groups.

Item	Unaffected Farmer Group	Affected Farmer Group	t-Test
Area (ha)	0.68	0.68	−0.01
	(1.69)	(1.34)
Yield (kg per ha)	4771	4407	5.55 ***
	(1362)	(1607)
Revenue (EUR per ha)	1237.1	1131.0	5.40 ***
	(419.3)	(454.0)
Total cost (EUR per ha)	568.7	543.3	1.71 **
	(329.2)	(309.3)
Seeds cost (EUR per ha)	85.1	95.5	−3.06 ***
	(75.3)	(70.5)
Chemical fertilizer cost (EUR per ha)	200.4	200.0	0.05
	(162.3)	(154.4)
Organic fertilizer cost (EUR per ha)	31.0	33.7	−0.86
	(70.1)	(56.4)
Pesticide cost (EUR per ha)	63.0	60.6	0.89
	(57.0)	(54.8)
Outsourced labor cost (EUR per ha)	72.2	62.6	1.87 **
	(112.5)	(107.4)
Hiring machine cost (EUR per ha)	116.9	90.7	4.50 ***
	(132.3)	(112.4)
Net revenue (EUR per ha)	668.4	587.7	4.85 ***
	(361.5)	(367.1)
Household labor (days per ha)	160.1	164.9	−0.91
	(117.9)	(108.6)
Reserve rice (kg per year)	471	638	−5.12 ***
	(546)	(1019)

**, *** are significant levels at 5% and 1%, respectively. The numbers in parentheses are the standard deviation.

presents a comparison of rice production characteristics between two farmer groups, and a t-test was conducted to analyze the mean differences. Both groups cultivated an average rice area of 0.68 ha. Natural disasters and pest infestations had a negative impact on the affected group’s average rice yield, which was 4407 kg per hectare as opposed to 4771 kg per hectare for the unaffected group. As a result, the unaffected group exhibited higher revenue and profitability.

The group of unaffected farmers had higher total average production costs compared to the group of affected farmers. Specifically, the unaffected group paid more for outsourced labor and hiring machinery. Notably, the affected group spent more on seed costs. Prolonged rain results in waterlogging of newly sown rice fields, necessitating farmers to resow and consequently increasing seed costs. Additionally, Vietnamese farmers traditionally sow rice with high density to mitigate losses caused by birds, mice, and yellow snails, which negatively impact seed survival rates. No significant differences were observed between the two groups in terms of costs associated with chemical fertilizers, organic fertilizers, and pesticides.

Moreover, farmers affected by natural disasters and pest infestations tended to store more rice compared to those in the unaffected group. This may be attributed to the uncertainties associated with natural disasters and the desire to ensure an adequate food supply during challenging times.

3.2. Technical Efficiency in Rice Production

Table 7. The results of the SFA model.

Variable	Coefficient	Standard-Error	t-Ratio
Beta 0	8.004 ***	0.036	221.889
Seeds	0.074 ***	0.007	11.037
Chemical fertilizers	0.056 ***	0.005	12.095
Organic fertilizers	0.009 ***	0.002	3.884
Pesticides	0.027 ***	0.004	6.357
Household labor	−0.020 ***	0.007	−2.972
Outsourced labor	0.009 ***	0.002	3.807
Machinery	0.011 ***	0.002	4.555
Sigma-square	0.533 ***	0.086	6.164
Gamma	0.954 ***	0.008	126.059
Log likelihood function	−230.738
LR test of the one-sided error	743.828

*** is significant level at 1%.

presents the results of the SFA model, providing insights into the determinants of technical efficiency in rice production. The findings highlight the statistical significance of all input variables, including seeds, chemical fertilizers, organic fertilizers, pesticides, household labor, outsourced labor, and machinery, in influencing technical efficiency.

Notably, the majority of these input variables demonstrate a positive impact, suggesting that increasing their utilization has the potential to enhance technical efficiency in rice production, except household labor. However, the coefficients associated with these factors are relatively small. For instance, the seeds variable emerges as the most influential determinant, with a modest impact coefficient of only 0.074. This implies that a 1% increase in seed usage corresponds to a mere 0.074% increase in technical efficiency. Similarly, although the other input variables are statistically significant, their impacts on technical efficiency are relatively modest.

A noteworthy finding is the negative impact of household labor on technical efficiency. This implies that relying on household labor in rice production is associated with lower technical efficiency. This could be attributed to various factors, such as inadequate skills or training for household laborers in rice production. In the context of Vietnam, rice-farming households may often take advantage of family labor, such as the elderly, women, or children, for rice production.

Furthermore, the Gamma value of 0.954, which is closer to 1, shows that inefficiency rather than random factors significantly influence the variability in rice output [28]. This finding further emphasizes the importance of addressing and improving technical inefficiencies to achieve enhanced productivity and technical efficiency in rice production.

Table 8. Technical efficiency (%) in rice production of the two farmer groups in each province.

Provinces	Whole Province	Unaffected Farmer Group	Affected Farmer Group	t-Test
Ha Tay	86.81	87.56	83.36	3.47 ***
	(8.18)	(7.59)	(9.84)
Lao Cai	74.97	77.96	70.33	4.77 ***
	(14.49)	(11.67)	(17.05)
Phu Tho	82.71	83.19	78.09	2.64 ***
	(10.14)	(9.02)	(17.33)
Lai Chau	70.37	72.85	65.10	4.52 ***
	(16.01)	(15.59)	(15.67)
Dien Bien	75.19	76.89	70.83	3.06 ***
	(17.15)	(15.28)	(20.67)
Nghe An	77.84	81.49	75.21	2.29 **
	(15.86)	(12.52)	(17.49)
Quang Nam	84.21	84.35	77.61	1.18
	(11.25)	(11.24)	(11.13)
Khanh Hoa	74.93	69.81	77.49	−1.31
	(15.26)	(12.47)	(16.16)
Dak Lak	83.29	82.83	84.02	−0.55
	(14.71)	(15.26)	(13.87)
Dak Nong	77.64	79.63	72.05	1.85 *
	(18.14)	(17.40)	(19.35)
Lam Dong	58.33	61.60	56.87	0.52
	(14.57)	(20.74)	(12.22)
Long An	92.06	92.24	91.08	1.17
	(3.97)	(3.87)	(4.45)
Average	78.99	81.01	73.66	10.78 ***
	(15.25)	(13.63)	(17.81)

*, **, *** are significant levels at 10%, 5%, and 1%, respectively. The numbers in parentheses are the standard deviation.

shows the technical efficiency of rice production of the sampled farmers and contrasts it between two groups: farmers affected and unaffected by natural disasters and pest infestations in their rice production. The average technical efficiency in rice production among the study sample was found to be 78.99%. This result aligns closely with the technical efficiency rates reported in similar studies conducted in Vietnam and Cambodia. Particularly, a study on the technical efficiency of rice production in Vietnam using the Vietnam Household Living Standard Survey (VHLSS) 2006 dataset reported an efficiency rate of 81.6% [19]. Another study in Cambodia estimated the technical efficiency of rice production at 78.4% [15]. While acknowledging that the SFA calculation method makes it challenging to directly compare technical efficiency across different studies, such comparisons offer a relative perspective, illuminating the reliability of results within each study’s specific context.

The t-test results indicate a significant disparity in technical efficiency between the two groups. The unaffected farmer group exhibited a considerably higher average efficiency of 81.01%, whereas the affected group had a lower technical efficiency of 73.66%. These findings suggest that both farmer groups have room for improvement in terms of technical efficiency in rice production. The potential for enhancement is estimated to be 18.99% for unaffected farmers and 26.34% for affected farmers.

However, when comparing the technical efficiency of rice production between the two groups within each province, unaffected farmers do not consistently demonstrate higher levels of technical efficiency compared to affected farmers. In provinces such as Ha Tay, Lao Cai, Phu Tho, Lai Chau, Dien Bien, Nghe An, and Dak Nong, the group of farmers unaffected by natural disasters and pest infestations exhibited higher technical efficiency. Conversely, in provinces including Quang Nam, Khanh Hoa, Dak Lak, Lam Dong, and Long An, no difference in technical efficiency was observed between the two farmer groups.

The findings also highlight the variations in technical efficiency in rice production among farmers across provinces. Farmers in Long An province achieved the highest average technical efficiency of 92.06%. Provinces such as Ha Tay, Phu Tho, Quang Nam, and Dak Lak also displayed high levels of technical efficiency in rice production, surpassing 80%. On the other hand, rice farmers in Lam Dong province recorded the lowest technical efficiency at 58.33%. The remaining provinces had technical efficiency levels above 70%.

Table 9. Factors affecting technical inefficiency.

Variables	Coefficient	Standard-Error	t-Ratio
Delta 0	−1.811 ***	0.482	−3.759
Pest infestations	0.323 ***	0.071	4.548
Flood	0.683 ***	0.142	4.794
Drought	0.732 ***	0.132	5.547
Typhoon	0.519 ***	0.100	5.180
Ethnicity	−1.201 ***	0.207	−5.803
Gender	0.016	0.054	0.297
Age	−0.002 ***	0.000	−10.832
Education	−0.009	0.006	−1.511
Lao Cai	0.599 ***	0.175	3.421
Phu Tho	0.673 ***	0.173	3.880
Lai Chau	1.142 ***	0.237	4.814
Dien Bien	0.830 ***	0.201	4.139
Nghe An	1.239 ***	0.259	4.776
Quang Nam	0.840 ***	0.200	4.205
Khanh Hoa	1.585 ***	0.320	4.951
Dak Lak	0.020	0.133	0.149
Dak Nong	1.131 ***	0.253	4.468
Lam Dong	1.738 ***	0.341	5.096
Long An	−2.569 ***	0.723	−3.555
Sigma-square	0.533 ***	0.086	6.164
Gamma	0.954 ***	0.008	126.059
Log likelihood function	−230.738
LR test of the one-sided error	743.828

*** is significant level at 1%.

presents the influence of factors on technical inefficiency in rice production. The results show that four variables associated with natural disasters and pest infestations have a positive effect on technical inefficiency, indicating that these factors reduce the technical efficiency of rice production. Among these variables, droughts exhibit the most significant impact, with a coefficient of 0.732. Similar previous studies have also consistently demonstrated that drought exerts the most significant influence on the technical efficiency of rice cultivation in Cambodia [15] and Sri Lanka [16].

Floods and typhoons also demonstrate substantial coefficients of 0.683 and 0.519, respectively. The influence of pest infestations on technical inefficiency in rice production is relatively weaker, with an impact coefficient of 0.323. This is relatively understandable because pest infestations can be controlled through measures such as pesticide application and integrated pest management (IPM) practices [29,30]. Therefore, farmers are able to mitigate their negative impact on rice crops. In contrast, natural disasters, such as typhoons, floods, and droughts, are uncontrollable events. The intensity and occurrence of these events are unpredictable and pose more significant challenges to rice production [7].

Additionally, the findings highlight that variables such as ethnicity and age significantly affect technical inefficiency. Specifically, the ethnicity variable demonstrates the most notable negative impact on technical inefficiency, with a coefficient of −1.201. This suggests that Kinh farmers perform better in terms of technical efficiency in rice production compared to other ethnic minorities. This result is consistent with previous studies [19,31], emphasizing the importance of proactive government policies to support ethnic minorities in rice production as well as agricultural production. Addressing these disparities is essential for achieving inclusive and sustainable rice production practices and promoting social equity and resilience within the agricultural sector. Efforts should focus on providing equal access to resources, training, and technical assistance for ethnic minority farmers to bridge the gap in technical efficiency and enhance their participation in sustainable rice production. The age variable shows a modest negative impact (−0.002) on technical inefficiency, indicating that older farmers tend to exhibit higher efficiency in rice production. The variables of farmer’s education and gender did not yield statistically significant results in this study.

When analyzing the regional variables, it is observed that the majority of these variables have a significant impact on the technical inefficiency of rice production, except for in the Dak Lak province. Most variables positively contributed to technical inefficiency, indicating that these provinces have the potential to enhance the technical efficiency of rice production. Notably, the Long An provincial variable stands out with a negative effect on technical inefficiency, indicating a positive influence on the technical efficiency of rice production. Long An province belongs to the Mekong River Delta region, which is recognized as a prominent rice production hub in Vietnam. Soil quality, water availability, infrastructure, and institutional support play significant roles in determining rice farming outcomes at the regional level. Therefore, tailored approaches that consider these local conditions are necessary to enhance technical efficiency and promote sustainable rice production practices across Vietnam. Region-specific strategies could include promoting soil conservation practices, optimizing water management systems, and supporting the development of local networks for knowledge sharing and collaboration.

3.3. Farmer Responses to Losses in Rice Production

Table 10. Farmers’ responses to losses in rice production.

Responses	Solution 1		Solution 2
	Frequency	Percentage	Frequency	Percentage
Nothing	146	22.15	449	68.13
Reduced consumption	324	49.17	72	10.93
Sold land	4	0.61	1	0.15
Sold livestock	28	4.25	44	6.68
Sold other assets	2	0.30	1	0.15
Got assistance from relatives	19	2.88	14	2.12
Got assistance from government	4	0.61	1	0.15
Got assistance from NGOs	2	0.30	0	0.00
Borrowed money from bank	18	2.73	7	1.06
Borrowed money from others	7	1.06	7	1.06
Got insurance payment	6	0.91	1	0.15
Postponed payment of loans	1	0.15	0	0.00
Work more	6	0.91	7	1.06
Used savings	82	12.44	41	6.22
Others	10	1.52	14	2.12

presents the coping strategies employed by farmers following losses incurred from natural disasters and pest infestations in rice production. The data highlight a significant proportion of farmers’ limited capacity to cope with such losses. Specifically, among the surveyed farmers, 22.15% reported a lack of viable solutions to recover from the losses, while 68.13% relied on only a single coping strategy.

Farmers’ remedial measures to overcome losses from natural disasters and pest infestations during rice cultivation are predominantly passive. Remarkably, nearly half of the farmers (49.17%) resorted to reducing their consumption while awaiting income from the subsequent crop. This result aligns with previous research indicating that natural disasters reduce household income and expenditures [32]. However, this coping strategy may have negative consequences for food security and nutrition [33].

Only a small percentage (12.44%) of farmers had savings to prevent crop failures. Some farmers resorted to selling assets such as livestock (4.25%), land (0.61%), or other assets (0.30%) to address consumption challenges. Nonetheless, not every household possesses savings or assets that can be liquidated in order to resolve immediate problems [34]. Other coping strategies employed by farmers included seeking assistance from relatives and friends (2.88%), the government (0.61%), or NGOs (0.30%), as well as borrowing money from banks (2.73%) or other sources (1.06%). Measures such as credits, subsidies, transfers from relatives, and social assistance can aid households in recovering quickly from natural disasters [35]. Furthermore, timely government and international community intervention and support are crucial in helping farmers overcome the effects of natural disasters and restore their production in the short term [36]. Additionally, a minority of farmers chose to work longer hours per day (0.91%) or received compensation from insurance (0.91%). Policies such as crop insurance and social safety nets can bolster farmers’ resilience to disaster risks [37].

Climate change exacerbates natural disasters and pest infestations, leading to increased losses over time [9]. Strengthening the resilience of farmers is crucial to enabling effective recovery and enhancing their adaptive capacity in the face of these challenges. This calls for a comprehensive and integrated approach that prioritizes sustainability to effectively respond to these challenges [34,38,39]. Obviously, the statistical results show that not all farmers have measures to address the losses in rice production. Therefore, it is essential to provide timely support to farmers impacted by natural disasters and pest infestations in their rice production, particularly those who are economically disadvantaged. By enhancing farmers’ access to agricultural insurance, they can obtain adequate compensation for losses incurred in rice production, enabling them to respond promptly and effectively.

Our findings emphasize the importance of adopting sustainable agricultural practices and policies to mitigate the impacts of natural disasters and pests on rice production. Implementing climate-smart agricultural techniques such as precision agriculture, agroecology, and organic farming can improve resource efficiency, reduce environmental harm, and enhance resilience to climate variability [38]. Integrating pest management practices, including biological controls, crop rotation, and resistant cultivars, can minimize reliance on chemical pesticides and minimize ecological damage [14,29,30].

The government should establish early warning systems that provide timely alerts and information regarding natural disasters and pest outbreaks. This will assist farmers in taking proactive measures, such as adjusting crop schedules, to minimize losses [40]. In Vietnam, the rice sowing schedule is carefully tailored for each locality to “avoid planthoppers”, optimize pumping costs during the initial stages of the crop, and prevent end-of-crop flooding. Local agriculture extension stations play a pivotal role in achieving these objectives by closely monitoring planthopper developments within their respective areas and neighboring regions. Through the use of light traps, they conduct on-site observations and track the migration patterns of planthoppers. Additionally, these extension stations collaborate with hydrological stations to monitor water levels, facilitating the creation of a well-planned calendar for sowing rice crops. Given the varying intensity and duration of annual floods across different regions in Vietnam, the rice sowing schedule must consider these regional differences.

Moreover, the schedule is designed to ensure an isolation period between rice crops in the same rice field, effectively helping to interrupt the spread of pathogens. It is recommended that rice farmers in the same area synchronize their sowing activities. By avoiding prolonged periods of continuous sowing, they can prevent the accumulation and dissemination of pest infestations. Adhering to these practices makes the management of pest infestations more feasible, allowing for better care and cultivation of rice crops.

Additionally, improving water management systems, including irrigation infrastructure and water conservation techniques, becomes essential. These improvements will ensure a sufficient water supply for rice crops during droughts and help mitigate flood damage [10]. Proper water management plays a crucial role in safeguarding the livelihoods of farmers and the productivity of rice cultivation [41,42].

Furthermore, targeted policies and interventions are necessary to address the unique challenges faced by ethnic minority groups and farmers in different provinces, promoting inclusive and sustainable rice production practices. This requires collaboration among policymakers, researchers, extension services, and farmers’ associations to develop context-specific strategies.

4. Conclusions

The objective of this study was to estimate the technical efficiency of rice production among Vietnamese farmers and identify factors related to natural disasters and pest infestations that affect it using the SFA model. The study utilizes data from 2394 rice farmers in the VARHS 2018 dataset. The average technical efficiency is 78.99% under optimal conditions. Farmers affected by natural disasters and pests achieve 81.01% efficiency, while unaffected farmers achieve 73.66%.

The study finds that natural disasters and pest infestations, especially drought, decrease technical efficiency in rice production. Measures to manage these challenges can help improve efficiency. However, farmers have limited options to cope with losses. The Vietnamese government should provide support such as food assistance, resources (seeds, fertilizers, machinery), financial aid, and promotion of collaboration among agencies, associations, and organizations. Farmers should consider crop insurance and diversifying income sources.

Understanding these challenges is crucial for sustainable agriculture and improving farmers’ livelihoods. Effective disaster and pest risk management and appropriate measures can enhance technical efficiency, build resilience, and contribute to sustainable rice production in Vietnam.