Socio-economic and environmental barriers to increased agricultural production: New evidence from central Thailand

Abstract

Thailand’s agricultural production has played a major role in enhancing the sustainability of domestic food supplies and successful international exports. However, agricultural production and farm productivity remain relatively low, especially in rapidly ageing high out-migration areas and among small-scale farmers. In this study, we use new household survey data from Prachinburi Province to examine the probability of facing six specific barriers to agricultural production. Data were analysed using descriptive statistics and a series of logistic regressions to identify the determinants. The results show that age of household head, off-farm income, farm size, and farm type were all associated with the likelihood of different productivity barriers. Gender of household heads was found to be associated with the probability of farm labour shortage, with female-headed households being at a disadvantage. We recommend that policies should be formulated to address these significant factors in order to overcome or circumvent the different barriers and enhance the livelihoods of the local population.

Keywords

Agricultural production crop production systems rural development ageing logistic regression Thailand

Introduction

Southeast Asia is a major producer and exporter of rice and a significant exporter of aquaculture, and as a result, millions of households depend on agricultural production for their livelihoods (Hoang, 2020). Despite growing household productivity, farmers continue to face barriers when attempting to increase agricultural output and income. Specific challenges include climate variability, limited access to technology, lack of irrigation, degrading soil fertility, an ageing farmer population resulting in labour shortage, and price fluctuations (Dang et al., 2014). Despite the Green Revolution and subsequent technological advances, weather and climate remain the key factors determining agricultural productivity throughout Asia (Petcho et al., 2019; Tsusaka and Otsuka, 2013). Sub-optimal agricultural production can have significant profound effects on the composition of agricultural value chains, market access opportunities for small-scale farmers, and their collective organisation (Petcho et al., 2019).

Climate variability represents a significant challenge that influences choice and performance of agricultural production systems, and food security (Szabo et al., 2015, 2018). In Australia, the growing intensity and frequency of droughts has already had a significant impact on food production and food prices. This trend is likely to continue, as temperatures rise and rainfall becomes more erratic (Quiggin, 2007). Similar yield losses due to extended droughts are also predicted to occur in various locations in Europe (e.g. Reinermann et al., 2019). In 2016, France suffered from an unprecedented yield loss, which was attributed to unusual warm weather in late autumn months combined with an atypical wet spring the following year (Mbow et al., 2019). Likewise, yield loss in numerous crops has been predicted for North and Latin America (Lesk et al., 2016; Zampieri et al., 2017).

Although climate change can be considered an external factor, other intrinsic socio-demographic processes such as high rural to urban migration and related labour shortage can hamper agricultural production. Labour shortage is one of the most serious problems, not only because of a higher wage rate for hired labour, but also because of the massive migration of young people away from agriculture (de-Brauw, 2019). In Thailand, it has been estimated that about 12 million workers aged 15–34 years have left agriculture (Poungchompu et al., 2012), and as a result, there is an increasing number of older workers in agriculture (Faysse, 2019). As these elderly farmers retire, their farms are unlikely to be taken over by their sons and daughters, who tend to migrate to urban areas (Salaisook et al., 2020).

One of the major barriers to agricultural technology adoption is the low quality of agricultural inputs resulting from lack of information on quality inputs (Husen et al., 2017). In Thailand, the Royal Thai Government’s Thailand 4.0 policy is a new economic model based on technology-based and high-quality production (Bussi and Sameer, 2017). The main goal of this policy is to boost quality of life and to remove barriers associated with the middle-income trap. The policy agenda focuses on sustainable development through mechanisation, increased yields for agriculture and improving farmers’ life quality (Apinya, 2017). To achieve these ambitious goals, the government has been working to promote a change of farming patterns by improvements with ‘smart farming’ such as training the agricultural labour force to use technology for production and marketing (Aekarach, 2017).

Despite the importance of the implications of the above challenges, currently only limited quantitative research examines factors affecting the probability of different barriers to greater agricultural production in Thailand (Pokmontree et al., 2019). Based on the existing literature and our field observations, we hypothesise that factors, such as farm size, farm location, type of existing agricultural production, and household off-farm income affect the likelihood of facing all types of barriers to various degrees. By undertaking a comprehensive analysis, the present study will contribute not only to filling the current knowledge gaps but also to informing existing and forthcoming rural development policies, with the ultimate goal of increasing agricultural production and improving the livelihoods of local populations.

Materials and methods

The study area

Prachinburi Province (Figure 1) is located in the central part of Thailand between 13°39′-14°27′ N and 101°09′-102°07′ E. The total land area is 4,762.362 km² (476,236.16 hectares). Prachinburi is a productive agricultural region, benefiting from fertile clay soils and an extensive irrigation network. The main farming systems are irrigated rice production and pond aquaculture (Faysse and Philboon, 2019). In 2017, demand for jobs in the agricultural sector was estimated at 60,635 or 37.75% (Prachinburi Province Labour Office, 2017). The number of workers has dropped recently, and labour shortages are expected as domestic workers tend to move from agriculture to the industrial sector in order to find better paid work. At the same time, Prachinburi is a receiving destination for international migrants, especially from neighbouring countries.

Figure 1.

The study area.

Household survey

The farm household survey was conducted using a standardised and semi-structured questionnaire and a multi-stage sampling procedure. From four sub-districts in Prachinburi province, a total of 32 villages were selected, taking into account their different crop production systems. Using the minimum sample size suggested by Yamane’s (1973) procedure, 365 farm households were randomly selected, of which 128 were in Bang Taen, 73 in Bang Pla Ra, 116 in Dong Kra Tong Yam, and 46 in Bang Kung. The survey was conducted in August and October 2018. The interviews were conducted with the farm’s household head whenever possible. In absence of the household head, another main household member with experience in agricultural production was interviewed. A draft questionnaire was developed and pre-tested with 20 households outside the study area. The household’s socio-economic characteristics and agricultural information were collected as independent variables, while the dependent variables on six barriers were collected based on a 3-point Likert scale, which required the respondents to deliberate carefully.

Data analysis

Descriptive statistics including cross-tabulation, mean, standard deviation, frequency, and percentage were used to show the production level of the farms. The production of rain-fed and irrigated rice, fish, shrimp, and mixed production was calculated using Equation 1.

Y_{p} {= Q}_{y} {/C}_{a}

where Y_p denotes yield, Q_y denotes output in kilograms, and C_a denotes total cultivated area in rai (1 rai = 0.16 hectare) or cage volume in cubic metres.

The weighted average index (WAI) was used to assess the group-level aggregate of households’ perception of negative production impacts related to six specific barriers: climate variability; agricultural technology; farm investment; farm labour; agricultural marketing; and irrigation service. The values were calculated based on the three impact levels of barriers (low, moderate, and high). According to Miah (1993), the composite index for the set of barriers to agricultural production is formulated from Equation 2.

WAI = {fH(1.00) + fM(0.67) + fL(0.33)}/N

where WAI denotes weighted average index, fH denotes frequency of high levels, fM denotes frequency of moderate levels, fL denotes frequency of low levels, and N denotes the total number of observations. The coefficients 0.33, 0.67, and 1.00 are attached to the low, moderate, and high levels of perceived negative impacts, respectively. The WAI values in the range of 0.00–0.33, 0.34–0.66, and 0.67–1.00 are qualitatively interpreted as low, moderate, and high levels of negative impacts.

A logit regression model (e.g. Hajra et al., 2017; Szabo et al., 2016) was used to examine household-level barriers to increasing agricultural production as specified in Equation 3.

P_{i} = F (Z_{i}) = F (α + β X_{i}) = \frac{1}{1 + e^{- Z_{i}}} = \frac{1}{1 + e^{- (α + β X_{i})}}

where e represents the base of natural logarithm, which is approximately 2.718. P_i is the probability that the farm household faces the specific barriers. The three-level ordinal variables representing the barriers were converted into binary values by combining either the high and moderate levels or the moderate and low levels into one level. X_i is a set of explanatory variables that affect the likelihood that the farm household has faced barriers over the last 5 years, and β represents coefficients of the explanatory variables. Equation 3 can therefore be converted as Equation 4 (Pindyck and Rubinfeld, 1998).

{log}_{e} \frac{P_{i}}{1 - P_{i}} = Z_{i} = α + β X_{i}

The left-hand side of Equation (4) is the logarithm of odds that the farmer has faced each of the six barriers to agricultural production in the past five years. The definition and measurement of the variables included in the analysis are presented in Table 1. The gender of the household head is hypothesised to influence the likelihood of facing specific barriers to agricultural production. Male farmers typically face greater barriers to crop production than female farmers because female farmers are more environmentally conscious than male farmers (Burton, 2014). With regard to the age of the household head, farmers of different ages may display different proficiencies in the use of technology and have different levels of technology input (Minot et al., 2006). Age can also be a proxy indicator of a crop production system’s expertise and availability. Older household heads may be more experienced and knowledgeable about barriers to the crop production system than younger ones.

Table 1.

Model variables, their descriptions and measurement.^a

Variables	Description and measurement	Expected sign
Dependent variable
Climate variability	Barrier of climate variability over the past 5 years (1 = a major or moderate problem; 0 = otherwise)
Agricultural technology	Barrier of agricultural technology over the past 5 years (1 = a major problem; 0 = otherwise)
Farm investment	Barrier of financial service over the past 5 years (1 = a major problem; 0 = otherwise)
Farm labour	Barrier of farm labour over the past 5 years (1 = a major problem; 0 = otherwise)
Agricultural marketing	Barrier of agricultural marketing over the past 5 years (1 = a major problem; 0 = otherwise)
Irrigation service	Barrier of agricultural irrigation service over the past 5 years (1 = a major problem; 0 = otherwise)
Independent variables
Gender	Gender of household head (1 = male, 0 = female)	+
Age	Age of household head (1 = 60 years or older; 0 = under 60 years)	+
Education level	Highest education level of household head (1 = secondary and higher; 0 = otherwise)	+
Adults in family	Number of adults in family who assist on the farm (1 = 3 persons or more; 0 = otherwise)	+
Farming experience	Number of years of experience in farming (1 = higher experience; 0 = lower experience)	+
Farm size	Size of the farm (1 = 25 rai or more; 0 = otherwise)	+
Off-farm income	Farm household income obtained off the farm (Thai baht per year)	+/−
Household income	Annual household income (Thai baht per year)	+/−
Total production	Total output of production (kilograms)	+
Irrigated rice	Total output of irrigated rice production (kilograms)	+
Fish	Total output of fish production (kilograms)	+
Shrimp	Total output of shrimp production (kilograms)	+
Mixed type	Total output of mixed type production (kilograms)	+

^a1 USD = 29 THB as of December 2019.

Education is expected to be negatively associated with the likelihood of facing productivity barriers. Education often helps farmers to make informed decisions and to increase the propensity to collaborate with others. Education level in farm production has shown higher payoff for farm production (Poltasingh and Phanindra, 2018). The number of adults in the household is expected to be negatively associated with the likelihood of facing various barriers to crop production. The total number of adults in the household who assist on the farm constitutes the supply of family labour for production activities and can thus positively affect the process of crop production. Farm size can have a direct positive effect on the process of crop production. When production is small, it is likely that almost all food produced by the household is consumed. Nonetheless, production and income will differ depending on the relationship between the farm and the supply of non-farm labour (Kobe et al., 2017).

Results

Socio-economic characteristics of respondents

The survey results revealed that almost 70% of the households were male-headed, while 52% of the heads were under 60 years old (Table 2). The average age of household heads was 59. Accordingly, the farm households in the four sub-districts can be considered as belonging to older farmers. The majority of households (78%) had more than three adults in their household. It is accepted that more adults in the family will ensure a sufficient supply of labour for the farm production system. With regard to farm size, it was relatively large, ranging from 43 to 54 rai per household. The average agricultural production was 26,259 kg/household among the farmers affected by climate variability, 25,467 kg/household among those affected by agricultural technology, 26,259 kg/household by farm investment, 25,393 kg/household by farm labour, 29,322 kg/household by agricultural marketing, and 24,562 kg/household by irrigation service.

Table 2.

Descriptive statistics of respondents affected by the barriers. M = Mean, SD = Standard deviation, HH = Household, n = 365

Socio-economicVariables	Barriers to increase agricultural production
	Climate variability(n = 245)		Agri-technology(n = 246)		Agri-investment(n = 227)		Labour force(n = 253)		Agri-marketing(n = 125)		Irrigation service(n = 258)
	M or n	±SD or %	M or n	±SD or %	M or n	±SD or %	M or n	±SD or %	M or n	±SD or %	M or n	±SD or %
Gender of HH head
Male	163	67	163	66	156	69	179	71	91	73	174	67
Female	82	33	83	34	71	31	24	29	34	27	84	33
Age of HH head	58.0^M	±12.0^SD	59.X^M	±11.9^SD	59.0^M	±11.8^SD	59.6^M	±11.6^SD	57.5^M	±11.9^SD	58.2^M	±12.1^SD
60 years or under	133	54	127	52	117	52	125	50	68	54	137	53
60+ years	112	46	119	48	110	48	128	51	57	46	121	47
Education of HH head
Up to primary	183	75	183	74	168	74	190	75	95	76	194	75
Secondary	43	17	48	20	40	18	46	18	23	18	46	18
Higher	19	8	15	6	9	8	17	7	7	6	18	7
Adults in household	1.8^M	±1.1^SD	1.7^M	±1.1^SD	1.7^M	±1.0^SD	1.7^M	±1.1^SD	1.9^M	±1.0^SD	1.8^M	±1.1^SD
< 3 people	190	78	198	80	181	80	202	80	98	78	202	78
> = 3 people	55	22	48	20	46	20	51	20	27	22	56	22
Farm size	43.4^M	±35.6^SD	44^M	±36.7^SD	46.3^M	±36.6^SD	44.3^M	±35.9^SD	53.5^M	±43.2^SD	43.2^M	±35.4^SD
< 25 rai	82	33	85	35	65	29	83	33	31	25	86	33
> = 25 rai	163	67	161	65	162	71	170	67	94	75	172	67
Farming experience of HH head
< low	45	18	45	18	41	18	44	17	16	13	56	22
> = high	200	82	201	82	186	82	209	83	109	87	202	78
Agricutural production (kg/HH/year)	26,259^M	±29,325^SD	25,467^M	±27,006^SD	26,259^M	±29,325^SD	25,393^M	±26,880^SD	29,322^M	±31,490^SD	24,562^M	±26,572^SD
Rainfed rice	1,513^M	±5,194^SD	2,373^M	±7,189^SD	2,082^M	±6,762^SD	2,315^M	±7,115^SD	2,160^M	±7,676^SD	2,208^M	±6,966^SD
Irrigated rice	21,013^M	±24,862^SD	29,585^M	±24,803^SD	20,899^M	±25,914^SD	19,886^M	±24,813^SD	25,772^M	±30,304^SD	19,186^M	±24,333^SD
Fish	83^M	±545^SD	44^M	±411^SD	156^M	±806^SD	267^M	±1,713^SD	28^M	±171^SD	197^M	±1,152^SD
Shrimp	179^M	±877^SD	112^M	±556^SD	227^M	±1,211^SD	239^M	±1,188^SD	11^M	±86^SD	328^M	±1,772^SD
Mixed production	2,516^M	±9,851^SD	2,354^M	±8,994^SD	2,894^M	±10,664^SD	2,685^M	±9,138^SD	1,351^M	±7,944^SD	2,641^M	±9,076^SD

Barriers to increased agricultural production

A summary of the weighted average index (WAI) for barriers to increasing different types of agricultural production for small-scale and large-scale farms is shown in Figure 2 and Figure 3, respectively. For small-scale farmland, the most critical barriers to increased production were climate variability for mixed products and agricultural technology for fish aquaculture, exhibiting a value of 0.7. The farm labour barrier had a value of 0.6 for fish, shrimp production, and mixed products. For rice production, the farmers perceived that all six barriers constituted moderate-level obstacles to increased production.

Figure 2.

Barriers to increased agricultural production for small-scale farms.

Figure 3.

Barriers to increased agricultural production for large-scale farms.

For large-scale farmers, the critical barriers were agricultural technology for mixed products, and irrigation services for shrimp aquaculture, both having a value of 0.8. Climate variability for mixed products and fish aquaculture and labour shortage for shrimp aquaculture were also regarded as serious constraints. Farmers perceived that rice production faced minor constraints, while fish, shrimp, and mixed products farming faced greater obstacles to increased production.

Results of the six logit regressions of barriers to increased agricultural production are shown in Table 3. In the climate variability model (Model 1), two of the 13 independent variables were significant in explaining perceptions of the households’ likelihood of facing climate variability to an increase in agricultural production. Off-farm income and irrigated rice yield were positively associated with this barrier. The insignificance of most variables in this model could be related to the greater importance of omitted factors, such as access to weather forecast and marketing information (Abid et al., 2018).

Table 3.

Factors influencing the probability of facing barriers to greater agricultural production: logit regression

Variables	Model 1Climate variability				Model 2Agri-technology				Model 3Farm investment				Model 4Farm labour				Model 5Agri-marketing				Model 6Irrigation service
Variables	Coeff.	Standarderror	p-value	Odds ratio	Coeff.	Standarderror	p-value	Odds ratio	Coeff.	Standarderror	p-value	Odds ratio	Coeff.	Standarderror	p-value	Odds ratio	Coeff.	Standarderror	p-value	Odds ratio	Coeff.	Standarderror	p-value	Odds ratio
Socio-economic
Gender	−0.3442	0.4944	0.204	0.7088	0.2610	−0.1536	0.614	0.8577	0.1622	0.3057	0.533	1.1761	0.4944	0.4596	0.078*	1.6395	0.2985	0.3691	0.276	1.3478	0.0396	0.2906	0.887	1.0404
Age	0.0013	0.0677	0.934	1.0013	0.0186	0.0148	0.418	1.0149	0.0457	0.0159	0.003***	1.0467	0.0677	0.0181	0.000***	1.0700	−0.0164	0.0161	0.318	0.9838	0.0325	0.0172	0.051**	1.0330
Education	0.1641	0.2624	0.488	1.1783	0.2957	0.0731	0.790	1.0759	0.4644	0.3699	0.046**	1.5910	0.2624	0.3139	0.277	1.3001	−0.0639	0.2305	0.795	0.9381	−0.0020	0.2439	0.993	0.9980
Adults in family	−0.2086	−0.3060	0.111	0.8117	0.1034	−0.4133	0.008***	0.6615	−0.2252	0.1012	0.075*	0.7984	−0.3060	0.1035	0.030**	0.7364	0.1605	0.1603	0.240	1.1741	−0.1088	0.1212	0.421	0.8969
Off-farm income	0.0000	0.0001	0.027**	1.0000	0.0000	0.0002	0.000***	1.0002	0.0001	0.0000	0.006***	1.0001	0.0001	0.0000	0.002***	1.0001	0.0000	0.0000	0.901	1.0000	0.0001	0.0000	0.001***	1.0001
Annual income	0.0000	0.0000	0.741	1.0000	0.0000	−0.0000	0.000***	1.0000	0.0000	0.0000	0.015**	1.0000	0.0000	0.0000	0.088*	1.0000	0.0000	0.0000	0.031**	1.0000	0.0000	0.0000	0.013**	1.0000
Agricultural production
Farm size	−0.0088	0.0020	0.203	0.9912	0.0081	−0.0064	0.430	0.9936	0.0143	0.0071	0.041**	1.0144	0.0020	0.0071	0.774	1.0020	0.0062	0.0073	0.398	1.0062	−0.0018	0.0070	0.802	0.9983
Farming experience	0.004	−0.0247	0.714	1.004	0.0129	−0.017	0.177	0.9825	−0.0194	0.0107	0.074*	0.9808	−0.0247	0.0120	0.044**	0.9756	0.0194	0.0115	0.085*	1.0196	−0.0349	0.0117	0.004***	0.9657
Rainfed rice	0.0000	0.0001	0.870	1.0000	0.0001	0.0001	0.011**	1.0001	0.0000	0.0000	0.728	1.0000	0.0001	0.0001	0.019**	1.0001	0.0000	0.0000	0.201	1.0000	0.0001	0.0000	0.021**	1.0001
Irrigation rice	0.0001	0.0000	0.000***	1.0001	0.0000	0.0001	0.000***	1.0001	0.0000	0.0000	0.072*	1.0000	0.0000	0.0000	0.002***	1.0000	0.0000	0.0000	0.036**	1.0000	0.0000	0.0000	0.004***	1.0000
Fish	−0.0003	−0.0001	0.132	0.9997	0.0003	−0.0003	0.285	0.9997	−0.0001	0.0001	0.109	0.9999	−0.0001	0.0001	0.250	0.9999	−0.0003	0.0003	0.392	0.9997	−0.0001	0.0001	0.116	0.9999
Shrimp	0.0000	0.0001	0.831	1.0000	0.0001	0.0000	0.743	1.0000	0.0000	0.0001	0.655	1.0000	0.0001	0.0001	0.382	1.0001	−0.0015	0.0009	0.073*	0.9985	0.0002	0.0001	0.090*	1.0002
Mixed agriculture products	0.000X	−0.0000	0.873	1.0000	0.0000	−0.0000	0.036**	1.0000	0.0000	0.0000	0.797	1.0000	−0.0000	0.0000	0.135	1.0000	−0.0000	0.0000	0.101	1.0000	0.0000	0.0000	0.186	1.0000
Constant	0.619X	−2.8304	0.501	1.857X	3.330X	1.1257	0.297	3.0823	−1.9474	0.1271	0.029**	0.1427	−2.8304	0.0563	0.003****	0.0590	−0.73456	0.4703X	0.454	0.4797	−0.0630	0.8937	0.947	0.9389
Likelihood ratio statistic	−197.255	−180.931			−161.556				−207.232				−180.931				−194.223				−188.801
P-value	0.000***	0.000***			0.000***				0.000***				0.000***				0.000***				0.000***
Pseudo R-squared	0.147	0.196			0.300				0.144				0.196				0.172				0.145
Num. of observations	245	253			246				227				253				125				258

Note: ***, **, and * denotes significant at 1.0%, 5.0%, and 10.0% respectively

In the agricultural technology model (Model 2), six of the 13 independent variables were significant in explaining households’ chances of facing the technological barriers. In particular, off-farm income, annual income, and farming both rain-fed and irrigated rice were positively associated with the likelihood of limited access to agricultural technology. On the other hand, the number of adults in a family and mixed production practices negatively influenced the likelihood of facing this barrier. The farm labour model (Model 3) showed that seven of the 13 independent variables were significant. Male authority, age of household head, off-farm income, and rainfed rice production positively influenced the likelihood of facing the barrier relating to farm labour, while the number of adults in the family, annual income, and farming experience had negative influences.

Regarding the agricultural marketing model (Model 4), four of the 13 independent variables were significant. Factors positively associated with the likelihood of facing this barrier included longer farming experience and irrigated rice yield. The number of adults in the family and annual income had negative influences on the barrier relating to agricultural marketing. The farm investment model (Model 5) revealed that age of household head, education of household head, off-farm income, farm size, and irrigated rice yield had a positive influence on the likelihood of facing difficulties in farm investment. In contrast, the number of adults in the family, greater annual income, and longer farming experience were found to be negatively associated with the likelihood of facing this barrier.

Finally, the irrigation service model (Model 6) showed that age of household head, off-farm income, farm size, farming both rainfed and irrigated rice, and shrimp production were all significantly and positively influenced the likelihood of difficulties in accessing irrigation services. Greater annual income and longer farming experience were found to reduce the likelihood of this barrier.

Discussion

Our results provide new insights into the socio-economic and environmental barriers to increased agricultural production in central Thailand. To the best of our knowledge, this is the first quantitative analysis of this type in the recent years and in this region. Our findings imply that where an older household head was present barriers related to access to farm investment and labour were more likely to exist. This is mostly in line with the existing literature. For example, a study from rural China (Guo et al., 2015) found that the impact of shifts in the age of the farming population adversely affected agricultural investment. Consequently, different forms of training should be utilised to increase efforts for nurturing modern professional farmers, and policies should be implemented simultaneously in order to improve agricultural productivity. It was found that male-headed households were more likely to face farm labour shortage, which is consistent with Porst and Sakdapolrak (2018) who found that women in Thailand were less prone to urban migration and were making a significant contribution to field activities.

With regard to the impact of income on the probability of farmers facing agricultural production barriers, the results suggest that off-farm income may ensure cash flow and decrease income fluctuations. As a survival strategy, farmers often seek to diversify their sources of income and use other strategies to stabilise their earnings through off-farm activities (Petcho et al., 2019). Previous research by Gunawan et al. (2019) analysing the warehouse receipt system in Indonesia also showed that income had positive effects on adoption of agricultural innovations. The results also confirm the importance of farming experience. Other literature in this area (e.g. Aonngernthayakorn and Pongquan, 2017) confirms our findings and shows that more experienced farmers have a lower likelihood of facing specific agricultural production barriers. Consequently, increased expertise and asset accumulation may enable farming households to diversify their livelihoods through additional non-farm activities while maintaining food production, which leads to greater income and quality of life (Edirisinghe, 2015).

Our study showed that an increase in farmland size is associated with a higher probability of access to farm investment. This is likely to be related to economies of scale. For example, Pensupar and Oo (2015) assessed the agricultural extension programme implemented in selected townships in Thailand and argued that farmers could enjoy cost advantages by increasing size of farm, output, and scale of production. Uchook (2018) also discussed the Thai Government proposal to encourage farmers to follow a large-scale farming programme by dividing land into their own community and cultivating their own crops, which jointly lowers the cost of production and increases its competitiveness.

Agricultural technologies have essential competitive advantages that enhance the wellbeing of farmers (Wubneshe et al., 2019). However, addressing the shortage of labour in agriculture is critical for restoring capacity and preserving production growth (Salaisook et al., 2020). Another element also requiring consideration is that productivity barriers relating to climate variability can affect crop production but differ among crop production types (Ochieng et al., 2016). Although irrigated agricultural systems are expensive and require more investment and technology, improvements in yield and water use efficiency should not be overlooked (Asghar et al., 2018). Hence, farm credit deserves more attention and should be promoted.

Conclusion

In the context of Thailand’s 4.0 agenda, understanding agricultural production barriers and their determinants is important in terms of influencing policy development and encouraging farmers and local authorities to take full advantage of the instruments developed within the 4.0 agenda (Jones and Pimdee, 2017). Our empirical results have revealed that farmers cultivating irrigated rice are highly likely to face different agricultural production barriers compared to those farming rainfed rice. The barriers are primarily related to climate variability and agricultural technology adoption (Tsusaka and Otsuka, 2013). In contrast, pond aquaculture corresponded to a higher probability of barriers related to climate variability, agricultural technology, and a shortage of labour.

While our study advances knowledge in the area of barriers faced by farmers in relation to increasing agricultural production, it is not without limitations. The study focused on a specific area of Thailand (Prachinburi Province) and therefore, it may not be applicable universally. Yet, our findings are likely to be of relevance to similar situations, especially within Southeast Asia. The study also focuses on barriers to greater agricultural production, but not the factors essentially influencing agricultural production. Further studies would be required in order to investigate the factors specifically affecting the volume of agricultural production or output, as well as strategies to overcome any identified barriers.

Farmers, especially in rapidly developing countries like Thailand, must adapt to climate change seasonal variation in order to increase productivity, and adopt sustainable farming practices. Further understanding of the way barriers influence the choice of crop systems and encouragement to follow strategies for counteracting the adverse effects of those barriers are needed. The Thailand 4.0 agenda provided opportunities for farmers to develop and improve their skills. This can be achieved through establishing training activities, skills development, and capacity building, as well as specific policies and strategies aiming at alleviating productivity barriers. This would not only allow local farmers to improve yields, but would also improve their livelihoods and contribute to further local development.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Agence Nationale de la Recherche (French National Agency for Research) as part of the DOUBT project.

ORCID iD

Sylvia Szabo

References

Abid

Scheffran

Schneider

, et al. (2018) Farmer perceptions of climate change, observed trends and adaptation of agriculture in Pakistan. Environment Management 63(1):110–123.

Aekarach

(2017) Govt eyes support for startups. Bangkok Post. Available at: https://www.bangkokpost.com/tech/1219377/govt-eyes-support-for-startups (accessed 15 August 2020).

Aonngernthayakorn

Pongquan

(2017) Determinants of rice farmers’ utilization of agricultural information in central Thailand. Journal of Agricultural & Food Information 18(1):25–43.

Apinya

(2017) Prayut touts ‘Thailand 4.0’ for farmers: great chance to boost life quality, PM says. Bangkok Post. Available at: https://www.bangkokpost.com/thailand/general/1216821/prayut-touts-thailand-4-0-for-farmers (accessed 15 August 2020).

Asghar

Sasaki

Jourdain

, et al. (2018) Levels of technical, allocative, and groundwater use efficiency and the factors affecting the allocative efficiency of wheat farmers in Pakistan. Sustainability 10:1619.

Burton

(2014) The influence of farmer demographic characteristics on environmental behaviour: a review. Journal of Environmental Management 135:19–26.

Bussi

Sameer

(2017) Thailand 4.0 and the future of work. The Nation. Available at: https://www.nationthailand.com/opinion/30311956 (accessed 15 August 2020).

Dang

Bruwer

, et al. (2014) Farmers’ perceptions of climate variability and barriers to adaption: lessons learned from and exploratory study in Vietnam. Mitigation and Adaptation Strategies for Global Change 19(5):531–548.

de-Brauw

(2019) Rural Youth: Determinants of Migration Throughout the World. Rural Development Report No. 15: Background Papers. Rome: International Fund for Agricultural Development.

10.

Edirisinghe

(2015) Smallholder farmers’ household wealth and livelihood choices in developing countries: a Sri Lankan case study. Economic Analysis and Policy 45:33–38.

11.

Faysse

(2019) Renewing the engagement of young people in farming in Thailand: possible pathways. Outlook on Agriculture 48(4):271–272.

12.

Faysse

Philboon

(2019) Building capacities for change: farmers moving away from convention rice farming in Prachinburi province, Thailand. Journal of Community Development Research 12(4):57–67.

13.

Gunawan

Kuwornu

JKM

Datta

, et al. (2019) Farmers’ perception of the warehouse receipt system in Indonesia. Sustainability 11(1690):2–21.

14.

Guo

Wen

Zhu

(2015) The impact of ageing agricultural labor population on farmland output: from the perspective of farmer preferences. Mathematical Problems in Engineering 2015(4):1–7.

15.

Hajra

Szabo

Tessler

, et al. (2017) Unravelling the association between the impact of natural hazards and household poverty: evidence from the Indian Sundarban Delta. Sustainability Science 12(3):453–464.

16.

Hoang

(2020) Investigating the agricultural competitiveness of ASEAN countries. Journal of Economic Studies 47(2):307–332.

17.

Husen

Loos

Siddig

KHA

, et al. (2017). Social capital and agricultural technology adoption among Ethiopian farmers. American Journal of Rural Development 5(3): 65–72.

18.

Jones

Pimdee

(2017) Innovative ideas: Thailand 4.0 and the Fourth Industrial Revolution. Asian International Journal of Social Sciences 17(1):4–35.

19.

Kobe

Olamide

Bamidele

, et al. (2017) Economic assessment of agricultural land market in rural Nigeria: pattern and drivers. Journal of Land and Rural Studies 6(1):50–66.

20.

Lesk

Rowhani

Ramankutty

(2016) Influence of extreme weather disasters on global crop production. Nature 529:84–87.

21.

Mbow

Rosenzweig

Barioni

, et al. (2019) Food security. In: Shukla

Skea

Calvo

(eds) Climate Change and Land: An IPCC Special Report on Climate Change, Desertification, Land Degradation, Sustainable Land Management, Food Security, and Greenhouse Gas Fluxes in Terrestrial Ecosystems. Rome: IPPC, p. 435.

22.

Miah

(1993) Applied Statistics for Social and Management Sciences. Singapore: Springer.

23.

Minot

Eoorecht

Anh

TTT

, et al. (2006). Income diversification and poverty in northern upland of vietnam. Available at: https://doi.org/10.2499/0896291480 (accessed 15 August 2020).

24.

Ochieng

Lilian

Mathenge

(2016) Effects of climate variability and change on agricultural production: the case of small scale farmers in Kenya. NJAS-Wageningen Journal of Life Science 77:71–78.

25.

Pensupar

(2015) The concept of economy of scale in agricultural extension program of Thailand. Asia-Pacific Information Platform on Agricultural Policy. Available at: http://ap.fftc.agnet.org/ap_db.php?id=526&print=1 (accessed 12 July 2020).

26.

Petcho

Szabo

Kusakabe

, et al. (2019) Farmers’ perception and drivers of membership in rice production community enterprises: evidence from the central region, Thailand. Sustainability 11:1–17.

27.

Pindyck

Rubinfeld

(1998) Econometric Models and Economic Forecasts, 4th ed. New York, NY: McGraw-Hill.

28.

Pokmontree

Prasitdamrong

Jermsittiparsert

(2019) Exploring the barriers and readiness factors for agricultural extension in Thailand. International Journal of Innovation, Creativity and Change 7(2):160–177.

29.

Poltasingh

Phanindra

(2018) Impact of farmer education on farm productivity under varying technology: case of paddy growers in India. Agricultural and Food Economics 6(7):1–19.

30.

Porst

Sakdapolrak

(2018) Advancing adaptation or producing precarity? The role of rural-urban migration and translocal embeddedness in navigating household resilience in Thailand. Geoforum 97:35–45.

31.

Poungchompu S, Tsuneo

Poungchompu

(2012) Aspects of the aging farming population and food security in agriculture for Thailand and Japan. IJERD – International Journal of Environmental and Rural Development 3(1):102–107.

32.

Prachinburi Province Labour Office (2017) Situation and Labour Indicators Index in Prachinburi 1st Quarter (January–March). In Thai. Prachinburi province.

33.

Quiggin

(2007) Complexity, climate change and the precautionary principle. Climate Change Working Paper: CO7#3, Brisbane, Australia.

34.

Reinermann

Gessner

Asam

, et al. (2019) The effects of droughts on vegetation condition in Germany: an analysis based on two decades of satellite earth observation time series and crop yield statistics. Remote Sensing 11(15):1783–1804.

35.

Salaisook

Faysse

Tsusaka

(2020) Reasons for adoption of sustainable land management practices in a changing context: a mixed approach in Thailand. Land Use Policy 96:104676.

36.

Szabo

Hossain

Fabrice

, et al. (2018). Accelerating progress towards the Zero Hunger Goal in cross2 boundary climate change hotspots. Environment: Science and Policy for Sustainable Development 60(3): 18–27.

37.

Szabo

Hossain

MDS

Adger

, et al. (2016) Soil salinity, household wealth and food insecurity in tropical deltas: evidence from south-west coast of Bangladesh. Sustainability Science 11(3):411–421.

38.

Szabo

Hossain

Matthews

, et al. (2015). Soil salinity, household wealth and food insecurity in agriculture-dominated delta. Sustainability Science 11(3): 411–421.

39.

Tsusaka

Otsuka

(2013) The declining impacts of climate on crop yields during the Green Revolution in India: 1972 to 2002. In: Otsuka

Larson

(eds) An African Green Revolution: Finding Ways to Boost Productivity on Small Farms. Dordrecht: Springer, pp. 71–94.

40.

Uchook

(2018) Thailand’s large-scale farming model: problem and concerns. FFTC Agricultural Policy Platform (FFTC-AP). Available at: http://ap.fftc.agnet.org/index.php (accessed 1 July 2020).

41.

Wubneshe

Zeller

Loos

(2019) The impact of agricultural technologies on poverty and vulnerability of smallholders in Ethiopia: a panel data analysis. Social Indicators Research (Open Access) 147:1–28.

42.

Yamane

(1973) Statistics: An Introduction Analysis. New York: Harper and Row.

43.

Zampieri

Ceglar

Dentener

, et al. (2017) Wheat yield loss attributable to heat waves, drought and water excess at the global, national and subnational scales. Environmental Research Letters 12(6):064008.