Livelihood Capital, Speculative Preferences, and Natural Rubber Farmers’ Participation in Cooperatives

Abstract

The development of cooperatives represents an effective solution to address the looming issue of “who will harvest the rubber”. Participation in cooperatives has the potential to increase the income of natural rubber farmers, enhance agricultural operational efficiency, and mitigate risks inherent in agricultural production. Livelihood capital and speculative preferences are key factors influencing natural rubber farmers’ decisions to participate in cooperatives. However, the existing literature has largely overlooked the intrinsic relationship between livelihood capital, speculative preferences, and the participation of natural rubber farmers in cooperatives. This study employs data from a field survey of 506 natural rubber farmers in Hainan Province, utilizing a Logit model to assess the impact of livelihood capital on farmers’ participation in cooperatives. The results indicate that (1) Livelihood capital encourages natural rubber farmers’ participation in cooperatives at the 5% significance level; (2) Speculative preferences negatively moderate the effect of livelihood capital on farmers’ participation in cooperatives. Therefore, the government can enhance farmers’ livelihood capital through education and training while providing financial instruments, such as insurance, to reduce speculative demand, thereby encouraging their participation in cooperatives.

1. Introduction

The development of farmer cooperatives is a crucial strategy for promoting rural revitalization and serves as the central mechanism for establishing a sustainable link between smallholder farming and modern agricultural development [1,2]. The 2024 Central Document No. 1 highlights the need to improve the operational and managerial capabilities of farmer cooperatives while also enhancing their capacity to support smallholder farmers. The Ministry of Agriculture and Rural Affairs has further emphasized the necessity of enhancing the operational levels of cooperatives, expanding their scale and influence, and improving their organizational efficiency and market competitiveness. The development of cooperatives presents an effective solution to address the forthcoming challenge of “who will harvest the rubber”. Rubber cooperatives, deeply embedded in rural areas and serving natural rubber farmers, play a pivotal role in enhancing farmers’ incomes [3]. In 2023, the total output of natural rubber in China’s three primary production regions—Yunnan, Hainan, and Guangdong—reached 895,000 tons, with approximately 1.36 million individuals engaged in rubber production and 536 cooperatives established [4]. Although progress has been made in the development of rubber cooperatives, challenges persist, including low participation among natural rubber farmers and weak cooperative structures, which limit their engagement in cooperatives. Therefore, investigating the factors influencing natural rubber farmers’ participation in cooperatives is essential for increasing their income, promoting rural revitalization, and advancing shared prosperity [5].

As a novel organizational model for agricultural management, cooperatives demonstrate characteristics such as extensive industry coverage, diverse organizational structures, comprehensive service functions, and a notable trend of consolidation [6]. Participation in cooperatives enables farmers to pool resources, reduce operational risks, and strengthen their market bargaining power [7,8]. Scholars have examined various aspects of farmer participation in cooperatives, focusing on social networks [9], emotional attachment [10], operational costs [11], and trust [12]. Additionally, the role of cooperatives in agricultural development has been extensively discussed, highlighting their positive impacts on increasing farmers’ incomes [8], adopting green technologies [13], reducing poverty [14], ensuring food security [3], and enhancing smallholder integration into agricultural and food value chains [15]. Existing research on cooperatives is relatively abundant; however, only a few studies have theoretically focused on natural rubber cooperatives. For instance, one study from Thailand examines the sustainable development issues of natural rubber cooperatives [16], while another from India highlights how communication problems affect the operations of these cooperatives [17]. There remains a notable gap in empirical research on how rubber farmers engage with natural rubber cooperatives.

Furthermore, drawing on the Sustainable Livelihoods Framework [18,19,20], the livelihood capital upon which farmers rely for their survival and long-term development plays a decisive role in shaping their behavior. In the context of natural rubber production, rubber farmers must thoroughly assess their economic conditions and resource endowments when deciding to participate in cooperatives. Their participation in cooperatives is influenced by multiple factors, with livelihood capital serving as a key determinant. However, there is a limited number of literature addressing the factors influencing natural rubber farmers’ participation in cooperatives. The mechanisms through which livelihood capital influences their participation in cooperatives remain poorly understood.

The previous literature has predominantly focused on the cooperative participation behavior of farmers involved in traditional staple crops [3,21]. However, rubber farmers face distinct production cycles, risk structures, and market environments, and the driving mechanisms behind their participation in cooperatives remain insufficiently explored. As a distinctive tropical economic crop [22], natural rubber has an extended planting cycle. Farmers typically derive no income during the first seven years prior to tapping the rubber trees, requiring substantial investments in labor, fertilizers, pesticides, and other production inputs [23]. The primary costs associated with natural rubber cultivation include land, seedling procurement, fertilizers, pesticides, tools, and labor management. As rubber trees are perennial, these cost components exhibit considerable flexibility. For example, fertilizers and pesticides can be applied minimally or omitted entirely, and maintenance inputs can be reduced as well. Consequently, natural rubber farmers incur high costs and face an extended period of no returns during the cultivation process.

In the prolonged practice of natural rubber production, the significant investments made by farmers often fail to yield the expected returns. Driven by a psychological aversion to future uncertainties, farmers frequently prefer using their livelihood capital for immediate gains rather than waiting for potentially higher future returns [24]. Speculative preference is defined as a tendency or behavioral inclination to engage in high-risk activities or investments with significant uncertainty to achieve excess returns [25].

According to behavioral economics theory, individuals with a higher preference for speculation tend to seek quick, short-term gains due to their higher tolerance for risk [26]. In contrast, individuals with a lower preference for speculation are more inclined toward long-term investments, seeking more stable returns [27]. This theoretical framework is further supported by empirical studies, particularly in financial investment decisions such as stock trading [28]. Such speculative preferences can influence rubber farmers’ perceptions of the long-term benefits of cooperatives, often leading them to prioritize immediate gains over the long-term advantages of collaboration.

Speculative preferences can undermine the positive impact of livelihood capital on rubber farmers’ participation in cooperatives. When speculative preferences are high, rubber farmers are more likely to be driven by short-term rewards, causing resistance to cooperatives, which require long-term commitment and stable returns. This diminishes the positive effect of livelihood capital on cooperative participation. In contrast, rubber farmers with lower speculative preferences are more likely to value long-term benefits. They are inclined to focus on the balance between long-term costs and returns and are more willing to invest their limited resources and livelihood capital into cooperative activities, even if the results take time to materialize.

Therefore, speculative preference serves as a moderating factor in the relationship between livelihood capital and natural rubber farmers’ participation in cooperatives. However, the manner in which speculative preference moderates this relationship remains inadequately explored in the existing literature. This study further investigates the underlying mechanism through which livelihood capital influences cooperative participation and examines the moderating effect of speculative preference within this context.

In summary, this study utilizes microdata collected from field surveys in Hainan Province and employs a Logit model to analyze the impact of livelihood capital on natural rubber farmers’ participation in cooperatives. The main contributions of this paper are as follows: First, based on the Sustainable Livelihood Framework; this study examines the effects of livelihood capital on natural rubber farmers’ participation in cooperatives through five dimensions: natural capital, physical capital, human capital, financial capital, and social capital. Second, this paper develops an analytical framework integrating livelihood capital, speculative preferences, and cooperative participation, providing an in-depth analysis of the moderating role of speculative preferences in the impact of livelihood capital on rubber farmers’ cooperative participation behavior. This study uniquely combines the Sustainable Livelihoods Framework with a behavioral economics perspective, revealing the interaction between livelihood capital and speculative preferences in shaping the participation of rubber growers in cooperatives. It fills a gap in the existing literature by addressing the overlooked psychological mechanisms behind short-term decision-making in the context of cash crop farmers. The findings of this research contribute not only to the exploration of socialized models for the natural rubber industry but also to the formulation and implementation of government policies related to cooperatives.

The structure of this study is organized as follows: Section 2 examines the theoretical foundation that underpins the research. Section 3 presents a detailed overview of the specific context of Hainan Province and the data sources employed. Section 4 describes the methodology and model framework used in the study. Section 5 reports the empirical analysis results. Section 6 provides a comprehensive discussion and reflection on the potential limitations of the research. Finally, Section 7 summarizes the findings and offers specific policy recommendations based on the results.

2. Theoretical Analyses

2.1. Livelihood Capital

Based on the Sustainable Livelihood Framework, livelihood capital refers to the resources or productive capacities possessed by natural rubber farmers [29]. It serves as a critical factor influencing their decision-making processes. Farmers with higher levels of livelihood capital are more likely to recognize the diverse benefits offered by cooperatives [7,8], including risk-sharing, access to market information, and substantial support in areas such as technical training and product marketing. This enhanced understanding motivates them to join cooperatives, seeking greater development opportunities and safeguards to mitigate uncertainties and risks in the production process.

Livelihood capital consists of five dimensions: natural capital, physical capital, human capital, financial capital, and social capital. Natural capital represents the endowment of natural resources available to natural rubber farmers. For instance, farmers with larger rubber plantations are more likely to leverage cooperatives to optimize production and resource utilization [30]. Farmers with weaker physical capital may struggle to navigate market fluctuations independently but can benefit from collective purchasing and sales through cooperatives, which help reduce production costs. Farmers with higher educational attainment possess greater human capital [19], equipping them with improved market awareness and the ability to adopt new technologies. These individuals are more likely to recognize and participate in cooperatives to capitalize on these advantages. Additionally, natural rubber farmers who have received training in rubber planting and tapping techniques [31] are better equipped to adopt advanced production methods. This not only enhances productivity but also instills confidence in joining cooperatives, where they can share expertise and benefit from additional technical support.

Financial capital, representing the funds available for production and reinvestment, also plays a role in influencing cooperative participation. Farmers with greater financial resources are more likely to engage in cooperatives, leveraging collective procurement and sales to gain a competitive market advantage. Social capital, defined by trust and collaborative relationships [9], facilitates farmers’ participation in cooperatives to secure better policy support and enhance competitiveness. The accumulation of livelihood capital improves farmers’ access to market information and their understanding of cooperatives, thereby promoting participation. Therefore, this study proposes the following hypothesis:

H1: 

Livelihood capital has a positive impact on natural rubber farmers’ participation in cooperatives.

2.2. Speculative Preferences

Speculative preference refers to natural rubber farmers’ tendency to prioritize short-term gains over long-term investment returns [25]. This preference may arise from farmers’ sensitivity to price fluctuations or an uncertain outlook on market trends, as opposed to securing long-term guarantees through sustained participation in cooperatives. When natural rubber farmers exhibit higher levels of speculative preference, they may be drawn to short-term profit opportunities, which can weaken the influence of livelihood capital on their decision to join cooperatives. Cooperatives generally advocate for shared cooperation and long-term development [3]. At the same time, individuals with speculative tendencies prioritize immediate, personal financial returns [27], which makes it difficult for them to align with the long-term goals of cooperatives. This is especially true in the context of frequent rubber price volatility [32], where speculative preferences are more likely to drive short-term behaviors, leading farmers to opt out of cooperative participation. In contrast, natural rubber farmers with lower speculative preferences tend to prioritize stable and enduring economic benefits over short-term speculative activities [33]. They are more likely to invest their livelihood capital in cooperative activities that may take longer to yield results [7]. Based on this, the following hypothesis is proposed (see Figure 1):

H2: 

Speculative preference plays a positive moderating role in the relationship between livelihood capital and natural rubber farmers’ participation in cooperatives.

3. Research Area and Data Sources

The data used in this study were collected through a field survey conducted between December 2021 and January 2022 by the National Rubber Industry Technology System, Industry Economics Post, in Hainan Province. The survey focused on rubber planting area, production volume, and geographic location [4] and included six counties in Hainan Province: Baisha, Chengmai, Danzhou, Wanning, Qiongzhong, and Qionghai, as shown in Figure 2. A stratified random sampling method was employed, initially selecting 2–4 towns from each county, followed by the random selection of 2–4 sample villages from each town, resulting in a total of 18 villages. In each village, 10–35 rural households were selected for face-to-face interviews. The survey employs a questionnaire-based approach, complemented by semi-structured interviews, to gain a comprehensive understanding of the rubber farmers’ situation [34]. The questionnaire addressed various aspects, including personal and household characteristics, rubber production conditions, ecological awareness, and satisfaction with rubber insurance. To ensure data quality, all members of the survey team received relevant training [35]. During the survey, a total of 550 questionnaires were administered across the 18 villages. After rigorous data cleaning, 506 valid questionnaires were retained, resulting in a response rate of 92%.

The characteristics of the sample are presented in

Table 1. Descriptive statistics of the sample farmers.

Items	Levels	Obs.	Frequency
Gender	Female	84	16.60%
	Male	422	83.40%
Age (years)	<30	8	1.59%
	31~40	62	12.25%
	41~50	147	29.05%
	51~60	194	38.34%
	>60	95	18.79%
Education (years)	Elementary and below	181	35.78%
	Junior high school	225	44.46%
	High school	86	17.00%
	University and above	14	2.77%
Ethnicity	Minority	280	55.34%
	Han	226	44.66%

. The natural rubber farmers in the Hainan sample were predominantly male, with 83.40% of participants identifying as male and 16.60% as female. To ensure more representative data, the study primarily selected rural residents who could communicate in Mandarin and whose responses reflected the views of their entire household. Most respondents were male, which is consistent with traditional gender roles in rural China, where “men work outside, and women manage the home”. Regarding age distribution, the largest group of natural rubber farmers was aged 50 to 60, accounting for 38.34% of the sample, while those under 30 were relatively few, comprising only 1.59%. This suggests that the rubber farming population is primarily composed of middle-aged individuals. In terms of ethnicity, 55.34% of natural rubber farmers were from ethnic minorities, while 44.66% were Han Chinese.

4. Model Construction and Variables

4.1. Model Construction

4.1.1. Logit Model

Since the dependent variable in this study, natural rubber farmers’ participation in cooperatives, is a binary discrete variable, the Probit model and Logit model are two primary methods within the binary choice model framework. One key assumption of the Probit model is that the random error term follows a normal distribution, which makes it highly dependent on the normality of the data in certain cases [36,37]. In contrast, the Logit model does not have such a normality requirement, offering greater flexibility in practical applications. Therefore, the Logit model was chosen for regression and econometric analysis [4]. The specific model used in this study is as follows:

In the model, $Y$ represents natural rubber farmers’ participation in cooperatives, $Capital$ denotes livelihood capital, and $X_i$ represents other factors influencing natural rubber farmers’ participation in cooperatives. ${\beta }_0$ is the intercept term, ${\beta }_i$ represents the estimated coefficients, and ${\mu }_i$ is the random error term as follows:

$$\log \mathrm{it}(Y)={\beta }_0+{\beta }_{1}Capital+{\displaystyle \sum {\beta }_i{X}_i}+{\mu }_i$$

(1)

4.1.2. Moderating Model

To examine the moderating effect of speculative preference in the relationship between livelihood capital and natural rubber farmers’ participation in cooperatives [4,11], the following model is employed:

$$\log \mathrm{it}(Y)={\beta }_0+{\beta }_{1}Capital+{\beta }_{2}SP+{\beta }_{3}Capital\ast SP+{\displaystyle \sum {\beta }_i{X}_i}+{\epsilon }_i$$

(2)

where $SP$ represents speculative preference, $Capital\ast SP$ denotes the interaction term between livelihood capital and speculative preference, ${\beta }_i$ represents the estimated coefficients, and ${\epsilon }_i$ is the random error term.

4.2. Variables

4.2.1. Dependent Variable

The dependent variable in this study is whether natural rubber farmers participate in a natural rubber cooperative, with participation denoted as 1 and non-participation as 0 [38].

4.2.2. Independent Variable

The core independent variable in this study is livelihood capital, which encompasses several dimensions. Natural capital represents the actual endowment of natural resources available to rubber farmers [30] and is measured using indicators such as altitude, the scale of natural rubber cultivation, and land fragmentation. Physical capital reflects the fixed assets that sustain the farmers’ livelihoods [31] and is measured by the number of agricultural machines, household appliances, and housing areas. Human capital reflects both the quantity and quality of labor in natural rubber production and is measured by the proportion of the labor force, the years of education of the farmers, and their involvement in technical promotion activities. Economic capital refers to the funds available for production [30,39] and is measured by natural rubber income, discretionary savings, and government subsidies. Social capital represents the social network resources that rubber farmers can leverage for their livelihood activities [40] and is measured using indicators such as the ability to borrow from others, the involvement of relatives or friends in the rubber business, and the extent of the farmers’ social networks.

The entropy method is used to calculate the level of livelihood capital for each rubber farmer and determine the weights for each dimension’s indicators. These weights are then applied to the respective dimensions of livelihood capital, yielding a composite measure of livelihood capital for each individual farmer as follows:

$$Capital={\displaystyle \sum _{i=1}^{5}W\ast }capita{l}_i$$

(3)

$Capital$ represents the livelihood capital evaluation value and $capita{l}_i$ refers to the individual dimensions of livelihood capital, including natural capital, physical capital, human capital, economic capital, and social capital. W denotes the weights of the indicators for each sub-dimension of livelihood capital. In the sustainable livelihood framework, the indicators for each sub-dimension are considered equally important; thus, a weight of 0.2 is assigned to each (

Table 2. Livelihood capital measurement indicators.

Primary Variable	Secondary Variable	Description
Natural capital	Altitude	The altitude of the respondent’s location (unit: meters)
	Rubber Scale	The scale of natural rubber cultivation (unit: mu)
	Land fragmentation	Land fragmentation = land scale/number of land plots
Physical capital	Agricultural Machinery	The number of agricultural machineries owned by the respondent, such as motorcycles and tricycles for transporting rubber (unit: pieces)
	housing area	The area of the respondent’s own housing (unit: square meters)
	Household Appliances	The number of household appliances owned, such as refrigerators and washing machines (unit: pieces)
Human capital	The proportion of labor force	The proportion of labor force in the family population
	Education	The actual number of years of education received by the respondent (unit: years)
	Participation in Promotion	Whether the respondent participates in the promotion of rubber technology, 0 = No; 1 = Yes
Economic capital	Rubber Income	Income from cultivating rubber (unit: yuan)
	Savings	The savings that the respondent can freely dispose of
	Government Subsidy	Subsidies from the government for cultivating rubber (unit: ten thousand yuan)
Social capital	Borrowing Ability	Whether the respondent can borrow money from relatives, friends, or others, 0 = No; 1 = Yes
	Rubber Business	Whether the respondent’s relatives or friends are engaged in the rubber business, 0 = No; 1 = Yes
	Social Network	How many people does the respondent often contact? 1 = Very few; 2 = Few; 3 = Average; 4 = Many; 5 = Very many

).

4.2.3. Moderating Variable

Building on the related literature [25], this study assesses natural rubber farmers’ speculative preferences by analyzing their behavior in purchasing lottery tickets. A specific question was included in the survey: “Do you purchase lottery tickets?”. If a rubber farmer responded “Yes”, it was inferred that their speculative preference was high, as purchasing lottery tickets is typically viewed as a pursuit of quick, high returns. In this case, speculative preference was assigned a value of 1. If the response was “No”, it was inferred that their speculative preference was low, as they were more likely to forgo immediate high returns in favor of less risky, long-term investments. Thus, purchasing lottery tickets was used as a proxy indicator to measure speculative preference in this study.

4.2.4. Control Variables

Drawing from the existing literature [8,12,19,40,41,42], this study selected gender, age, whether the farmer engages in multiple occupations, ethnicity, internet usage, disaster impact, information search behavior, years of production experience, participation in technical demonstration households, life satisfaction, and regional dummy variables as control variables. Descriptive statistics for the main variables are presented in

Table 3. Descriptive statistics of the variables.

Variable	Description	Min	Max	Mean
Dependent Variable
Participation in Cooperatives	Whether the rubber farmer participates in a natural rubber cooperative, with participation coded as 1 and non-participation coded as 0.	0	1	0.167
Independent Variable
Livelihood Capital	Derived from a combination of entropy and weight methods.	0.045	0.601	0.234
Moderating variable
Speculative Preferences	Whether you purchase lottery tickets? Not purchasing indicates low speculative preference, coded as 0, while purchasing indicates high speculative preference, coded as 1.	0	1	0.435
Control Variables
Gender	Gender of the respondent, 0 = Female; 1 = Male.	0	1	0.835
Age	Age of the respondent.	20	70	51.669
Multiple Occupations	whether the farmer engages in multiple occupations	0	1	0.675
Ethnicity	Ethnicity of the respondent, 0 = Han; 1 = Minority.	0	1	0.553
Internet Usage	Number of internet apps used by the respondent (WeChat, TikTok, and Kuaishou), 0 = None; 1 = One; 2 = Two; 3 = Three.	0	3	1.553
Disaster Impact	Whether you were affected by natural disasters such as typhoons or pests last year.	0	1	0.612
Information Search	Whether the respondent can use the internet to search for information.	0	1	0.478
Production Years	The actual number of years the respondent has been producing rubber.	0	40	20.193
Model Household	Whether the respondent is a local model household for rubber planting technology, 0 = No; 1 = Yes.	0	1	0.024
Life Satisfaction	Satisfaction with current life, 1 = Dissatisfied; 2 = Somewhat dissatisfied; 3 = Neutral; 4 = Fairly satisfied; 5 = Completely satisfied.	1	5	2.620
Central Region	Central region (Baisha and Qiongzhong) = 1, others = 0.	0	1	0.431
Eastern Region	Eastern region (Wanning and Qionghai) = 1, others = 0.	0	1	0.232

.

5. Results

5.1. Baseline Regression

Before conducting the baseline regression, this study used Stata17 software to perform a multicollinearity test on the variables. The VIF values ranged from 1.03 to 3.43, all of which are below 10, indicating that there is no multicollinearity issue with the selected variables. As shown in

Table 4. Baseline regression.

Variable	(1) Logit	(2) dy/dx	(3) Logit	(4) dy/dx
Livelihood capital	2.776 **	0.357 **
	(1.39)	(0.18)
Natural capital			−9.308 ***	−1.064 ***
			(2.71)	(0.30)
Physical capital			2.127 *	0.243 *
			(1.20)	(0.14)
Human capital			2.091 ***	0.239 ***
			(0.56	(0.06)
Economic capital			1.889	0.216
			(2.25)	(0.26)
Social capital			−0.361	−0.041
			(0.50)	(0.06)
Gender	−0.331	−0.042	−0.091	−0.01
	(0.32)	(0.04)	(0.35)	(0.04)
Age	0.024 *	0.003 *	0.027 *	0.003 *
	(0.01)	(0.00)	(0.02)	(0.00)
Multiple Occupations	−0.069	−0.009	−0.334	−0.038
	(0.27)	(0.03)	(0.29)	(0.03)
Ethnicity	0.433	0.056	0.806 **	0.092 **
	(0.27)	(0.03)	(0.32)	(0.04)
Internet Usage	−0.169	−0.022	−0.067	−0.008
	(0.16)	(0.02)	(0.18)	(0.02)
Disaster Impact	0.138	0.018	0.185	0.021
	(0.27)	(0.03)	(0.28)	(0.03)
Information Search	0.582 *	0.075 *	0.432	0.049
	(0.32)	(0.04)	(0.35)	(0.04)
Production Years	0.001	0.000	−0.007	−0.001
	(0.01)	(0.00)	(0.01)	(0.00)
Model Household	1.340 **	0.172 **	1.556 **	0.178 **
	(0.58)	(0.07)	(0.68)	(0.08)
Life Satisfaction	−0.374 ***	−0.048 ***	−0.320 **	−0.037 **
	(0.13)	(0.02)	(0.15)	(0.02)
Central Region	−0.18	−0.023	0.986 **	0.113 **
	(0.28)	(0.04)	(0.43)	(0.05)
Eastern Region	−1.190 ***	−0.153 ***	−1.619 ***	−0.185 ***
	(0.38)	(0.05)	(0.39)	(0.04)
Constants	−2.293 **		−2.411 *
	(1.07)		(1.25)
N	506	506	506	506
Wald chi2	36.82 ***		74.66 ***
Pseudo R2	0.083	0.083	0.180	0.180

Note: * indicates the level of statistical significance. *** p <  0.01, ** p <  0.05, * p < 0.1. Standard errors are in parentheses. Pseudo R2 is a measure of goodness of fit.

, total livelihood capital significantly influences natural rubber farmers’ participation in cooperatives at the 5% significance level. This indicates that livelihood capital positively promotes natural rubber farmers’ participation in cooperatives. The marginal effects derived from the logit model are reported in regressions (2) and (4) in Table 3. For example, regression (2) shows that for every 1% increase in livelihood capital, the likelihood of natural rubber farmers participating in cooperatives increases by 0.357. This can be explained by the fact that livelihood capital is a key determinant of cooperative participation among natural rubber farmers [40]. The greater the abundance of livelihood capital, the more likely farmers are to participate in cooperatives, as farmers with sufficient livelihood capital are generally better equipped to withstand market fluctuations and production risks. Participation in a cooperative provides an effective means of diversifying and reducing these risks [43]. Furthermore, farmers with abundant livelihood capital are more likely to access information and resources. Cooperatives, as platforms for sharing information and resources, enable farmers to exchange key information such as advanced rubber tapping techniques, market trends, and pest control strategies with other members.

Natural capital has a negative effect on farmers’ participation in cooperatives at the 1% significance level. This may be due to the fact that larger-scale farmers with more fragmented land holdings tend to incur higher production and transportation costs, which leaves them with less time and energy to engage in cooperative activities. In contrast, physical capital positively affects farmers’ participation in cooperatives. Abundant physical capital provides greater opportunities for production and market engagement [11]. By participating in cooperatives, farmers can achieve economies of scale, improve production efficiency, and ultimately enhance both their economic returns and social status [31]. Human capital also exerts a positive effect on farmers’ participation in cooperatives. Farmers with higher levels of human capital are more likely to engage in training and educational opportunities provided by the cooperative, which helps them continuously enhance their skills and knowledge.

Regarding control variables, older farmers, due to their extensive planting and management experience, are likely to have a more profound understanding of the numerous benefits offered by cooperatives, making them more inclined to participate [10]. Furthermore, farmers who are able to search for information online are less likely to experience information asymmetry and are better informed about the specific advantages of cooperatives, further encouraging their participation [44]. Farmers designated as technical demonstration households significantly contribute to promoting cooperative participation. These demonstration farmers, by participating in cooperatives, can access additional resources, such as production materials, market information, and technical support, which not only enhances their production capacity but also stimulates technological advancement and innovation within the cooperative [38]. Farmers with higher life satisfaction are less inclined to participate in cooperatives. A likely explanation is that these farmers typically have higher incomes and perceive their economic status as sufficiently secure, thereby not seeing the need to participate in cooperatives to improve their economic situation. At the regional level, farmers in the eastern regions exhibit a lower inclination to participate in cooperatives. The mountainous areas in the east face fewer production constraints and feature relatively flat land and convenient transportation, which enables farmers to operate more autonomously in their planting and sales, reducing their reliance on external resources, such as market support and technical services, provided by cooperatives.

5.2. Moderating Effect

This study further examines the moderating role of speculative preferences in the relationship between livelihood capital and rubber farmers’ participation in cooperatives.

Table 5. Moderating effect results.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
Livelihood Capital	6.614 ***
	(1.91)
Natural capital		−8.717 **
		(3.42)
Physical capital			3.880 ***
			(1.42)
Human capital				3.304 ***
				(0.70)
Economic capital					2.691
					(2.03)
Social capital						0.024
						(0.56)
Speculative Preferences	2.073 **	1.016	1.540 *	1.096	0.464	0.331
	(0.85)	(0.18)	(0.70)	(0.45)	(0.57)	(0.60)
Livelihood Capital × Speculative Preferences	−7.769 ***
	(2.65)
Natural capital × Speculative Preferences		−4.015
		(3.85)
Physical capital × Speculative Preferences			−4.569 **
			(2.30)
Human capital × Speculative Preferences				−2.192 **
				(0.96)
Economic capital × Speculative Preferences					−4.890
					(3.64)
Social capital × Speculative Preferences						−0.332
						(0.86)
Control variables	Controlled	Controlled	Controlled	Controlled	Controlled	Controlled
Constants	−3.098 ***	−1.355	−2.617 **	−2.880 **	−1.944 *	−1.045
	(1.07)	(1.11)	(1.11)	(1.09)	(1.03)	(0.98)
N	506	506	506	506	506	506
Wald chi2	44.68 ***	56.50 ***	36.20 ***	57.33 ***	34.37 ***	34.08 ***
Pseudo R2	0.099	0.142	0.089	0.128	0.077	0.075

Note: *** p < 0.01, ** p < 0.05, * p < 0.1. Standard errors are in parentheses. Pseudo R2 is a measure of goodness of fit.

shows that the interaction term between livelihood capital and speculative preferences has a regression coefficient of −7.769, which is statistically significant at the 1% level, indicating that speculative preferences negatively moderate the relationship between livelihood capital and rubber farmers’ cooperative participation. Specifically, rubber farmers with higher speculative preferences tend to prioritize short-term returns when making participation decisions [7,8], which leads them to be unwilling to invest their livelihood capital into long-term collaboration within cooperatives. In contrast, farmers with lower speculative preferences are more inclined to invest their livelihood capital in cooperatives, expecting long-term benefits.

While cooperatives emphasize resource sharing and long-term benefits, rubber farmers with high speculative preferences focus more on immediate economic gains rather than stable, long-term returns. This weakens the positive role of livelihood capital in promoting cooperative participation. Conversely, farmers with low speculative preferences are psychologically more inclined toward stable long-term returns [27], making them more willing to invest their livelihood capital in cooperatives to reap benefits such as market information, policy support, and technological improvements over time. These results suggest that speculative preferences, by influencing farmers’ expectations, significantly alter the positive effect of livelihood capital on cooperative participation.

Moreover, the regression analyses in

Table 6. Robustness regression results.

Variable	Substitutes the Measurement of the Independent Variable	1% Winsorization	OLS
Livelihood Capital	2.628 **	4.752 ***	0.436 **
	(1.21)	(1.52)	(0.19)
Constants	−5.197 ***	−4.981 ***	0.178
	(1.15)	(1.24)	(0.12)
Control variables	Controlled	Controlled	Controlled
N	506	485	506
R2/Pseudo R2	0.090	0.095	0.079

Note: *** p < 0.01, ** p < 0.05. Standard errors are in parentheses. Pseudo R2 is a measure of goodness of fit.

(columns 3 and 4) further reveal that speculative preferences mainly weaken the positive effects of material capital and human capital within livelihood capital on rubber farmers’ cooperative participation. A possible reason for this is that farmers with high speculative preferences lack long-term planning and a vision for future benefits, focusing more on immediate short-term returns. This short-term behavior leads them to invest more funds and resources into high-risk speculative activities, hoping for quick, high returns, which weakens the positive effect of material capital on cooperative participation. Additionally, rubber farmers with high speculative preferences tend to use their human capital for short-term profits rather than investing their livelihood capital in cooperative development. The role of human capital typically manifests through technical training, managerial experience, and long-term labor investment, all of which help enhance the efficiency and market competitiveness of cooperatives. This shortsighted behavior leads to a lack of interest in investing in long-term human capital and management experience required by cooperatives, thereby diminishing the positive effect of human capital on cooperative participation.

5.3. Robustness Test

This study utilizes both the entropy method and equal weighting method to measure livelihood capital. To assess the robustness of the baseline regression, the study also employs alternative variable measurement methods. Specifically, the weights for natural capital, social capital, material capital, financial capital, and human capital are assigned using the entropy method for each dimension of livelihood capital [45]. The results, presented in Table 6, indicate that, even after altering the variable measurement method, livelihood capital continues to significantly influence natural rubber farmers’ participation in cooperatives at the 5% significance level.

Additionally, the study applies a winsorization method, using 1% winsorization to all continuous variables. After winsorization, the effect of livelihood capital on farmers’ participation in cooperatives remains significant. Finally, ordinary least squares (OLS) regression is employed as a substitute for the logit model, and the results maintain robustness, further confirming the stability of the findings.

5.4. Heterogeneity Analysis

While the previous analysis examined the overall impact of livelihood capital on natural rubber farmers’ participation in cooperatives, this section investigates the potential heterogeneity of this relationship under different conditions.

Initially, the length of production experience influences natural rubber farmers’ decisions to join cooperatives [12]. The study categorizes farmers based on the median years of rubber tapping experience into two groups: those with shorter and longer production histories. As shown in

Table 7. Heterogeneity analysis results.

Variable	Shorter Production	Longer Production	No Disaster Impact	Disaster Impact	No Planting Structure Adjustment	Planting Structure Adjustment
Livelihood Capital	4.062 **	3.072	0.875	5.233 ***	6.897 ***	1.719
	(1.89)	(2.39)	(2.36)	(1.89)	(2.43)	(1.86)
Constants	−1.652	−9.127 ***	−2.303	−1.371	−3.893 **	−2.073
	(1.61)	(2.16)	(1.97)	(1.60)	(1.68)	(1.37)
Control variables	Controlled	Controlled	Controlled	Controlled	Controlled	Controlled
N	278	228	196	310	229	227
Pseudo R2	0.097	0.230	0.072	0.144	0.129	0.104

Note: *** p < 0.01, ** p < 0.05. Standard errors are in parentheses. Pseudo R2 is a measure of goodness of fit.

, for farmers with shorter production histories, livelihood capital positively impacts cooperative participation. This is primarily because less experienced farmers may lack market bargaining power and technical expertise, making them more willing to invest their livelihood capital in cooperatives to benefit from technical support and other services. In contrast, farmers with longer production histories are more accustomed to traditional production and marketing methods. They can independently access market information, manage risks, and enhance their technical skills, leading to a reduced dependency on the support provided by cooperatives. Therefore, livelihood capital does not significantly influence their decision to join cooperatives.

Secondly, natural disasters often drive natural rubber farmers to seek more effective risk management strategies [4], and cooperatives serve as a crucial mechanism for mitigating risks. The study divides farmers into two groups based on whether they experienced a natural disaster (e.g., typhoon) in the previous year. As shown in Table 7, for farmers who experienced a natural disaster, livelihood capital significantly positively influences their participation in cooperatives. This suggests that disaster-stricken farmers are more inclined to seek cooperatives as a platform for risk-sharing, resource pooling, and access to risk management strategies, thereby fostering resilience in the face of future shocks. On the other hand, for farmers who did not experience such shocks, livelihood capital does not significantly influence their cooperative participation. This is likely because these farmers perceive their existing resources and abilities as sufficient to manage everyday production and life, reducing their reliance on the supplementary support and services that cooperatives provide.

Lastly, during periods of long-term low rubber prices [39], natural rubber farmers may adjust their cropping structure by cultivating more economically viable crops, such as betel nuts. The study categorizes farmers into two groups: those who have adjusted their cropping structure and those who have not. As shown in Table 7, for farmers who have not adjusted their cropping structure, livelihood capital significantly encourages their participation in cooperatives. These farmers are more reliant on rubber production and require the risk management mechanisms and technical training that cooperatives offer. Livelihood capital helps them leverage these resources, thus increasing the likelihood of cooperative participation. Conversely, for farmers who have diversified their crops, livelihood capital does not significantly affect their decision to join cooperatives. This may be because these farmers have shifted their focus to crops such as betel nut or pepper, reducing their reliance on rubber production and, thus, their need for cooperative membership.

6. Discussion

The greater the livelihood capital of natural rubber farmers, the more likely they are to participate in cooperatives, a finding consistent with the existing literature [40,43,46]. An increase in livelihood capital enables natural rubber farmers to benefit from improved resource allocation, market access, and information flow through cooperatives. This not only enhances production efficiency but also strengthens their ability to manage market fluctuations and environmental risks. Specifically, abundant livelihood capital enables natural rubber farmers to fully utilize the technical support, training opportunities, and collective procurement services provided by cooperatives, thereby reducing production costs and increasing rubber yield [19]. Market conditions have a significant impact on rubber farmers’ decisions to participate in cooperatives [33]. Fluctuations in natural rubber market prices and the ability to access market information may drive rubber farmers to join cooperatives as a means of mitigating market risks [39].

Furthermore, joining a cooperative facilitates the scaling of rubber production, improving the consistency of latex quality, reducing costs in the processing stage, and enhancing raw material quality. This results in cost advantages and core competitiveness in rubber production, ultimately increasing production efficiency, market competitiveness, and farmers’ incomes.

However, due to the inherent vulnerabilities of natural rubber farmers and the potential influence of speculative preferences, they are often reluctant to invest resources in long-term benefits [33], resulting in a limited understanding of the advantages of cooperatives. This makes the process of enhancing livelihood capital long and complex [47]. In this context, guiding farmers to participate in cooperatives becomes particularly important. Rubber cooperatives not only serve as a means of improving the quality of farmers but also act as an important link between farmers and the government, helping to reduce information asymmetry and correct potential policy implementation biases.

Theoretically, this study contributes to the existing literature by confirming that livelihood capital promotes natural rubber farmers’ participation in cooperatives, while speculative preferences attenuate the positive effect of livelihood capital on cooperative involvement. Two key contributions are made: First, this study clarifies the role of various dimensions of livelihood capital in promoting cooperative participation, thereby broadening the scope of research on cooperative participation to encompass farmers’ livelihood capital. Second, livelihood capital and speculative preferences are significant factors shaping farmers’ behavior in cooperative participation. By adopting a psychological decision-making approach, this study explores how speculative preferences moderate the effect of livelihood capital on cooperative engagement, thereby enriching our understanding of the factors influencing cooperative participation.

This study has several limitations. First, Due to the lack of appropriate instrumental variables, this study does not address the issues of omitted variables and reverse causality between livelihood capital and rubber farmers’ participation in cooperatives. Future research could employ instrumental variables to tackle potential endogeneity problems further. Second, the study relies on cross-sectional data from a single year, which, although it provides a snapshot of natural rubber farmers’ cooperative participation at a specific point in time, fails to capture the continuity and evolution of their participation behaviors over time. Additionally, due to constraints such as time and financial resources, this study focuses on Hainan Province and does not include other natural rubber-producing regions. Therefore, future research should prioritize the collection and analysis of longitudinal data while expanding the sample area to achieve a more in-depth and comprehensive analysis. Third, given the limited body of research on speculative preferences, this study assesses speculative tendencies through a single indicator, which is whether farmers purchase lottery tickets. This approach may be overly simplistic, and future research could broaden the measurement of speculative preferences by incorporating additional indicators, such as the amount spent on lottery tickets, to provide more robust findings.

7. Conclusions and Implications

7.1. Conclusions

Livelihood capital is a critical factor driving natural rubber farmers’ participation in cooperatives. Rubber cooperatives, deeply embedded in rural communities and dedicated to serving natural rubber farmers, play a pivotal role in improving their incomes. This study, therefore, employs data from 506 field surveys conducted in Hainan Province and utilizes a Logit model to examine the impact of livelihood capital on natural rubber farmers’ participation in cooperatives. The results show that: (1) Livelihood capital fosters natural rubber farmers’ participation in cooperatives, with natural capital exerting a negative effect on participation, while material and human capital positively influence it. (2) Speculative preferences act as a moderator, negatively influencing the relationship between livelihood capital and cooperative participation. Speculative preferences mainly attenuate the positive influence of material and human capital on cooperative participation. (3) Group differences indicate that farmers with more extensive production experience, those who have faced disaster shocks, and those who have not modified their crop structure are more likely to benefit from livelihood capital in their cooperative participation.

7.2. Implications

Based on these findings, the study offers several policy recommendations. The government can implement a series of comprehensive measures to enhance natural rubber farmers’ livelihood capital, focusing on improving human and economic capital to meet the development needs of the rubber industry. First, educational and training programs should be introduced to enhance farmers’ professional skills and market competitiveness. Second, financial support and optimization of market mechanisms should be employed to increase farmers’ economic returns, thereby boosting their willingness to join cooperatives. Additionally, the government should collaborate with relevant organizations to provide financial instruments, mitigate speculative behavior, and explore risk management tools such as price insurance and risk funds [48]. Furthermore, policy incentives should be introduced to encourage natural rubber farmers to participate actively in cooperatives, including long-term low-interest loans, insurance subsidies, and technical support to enhance the attractiveness of cooperatives and foster long-term planning and investment. Finally, strong support for the development of rubber cooperatives is crucial, including the establishment of model cooperatives, the development of cooperative alliances, and the creation of demonstration effects to attract farmers to join cooperatives [4].