Abstract
Introduction
The function of adaptive capacity is contingent on interactions among multiple variables. Human, physical, social, financial, and natural capital must synergise for specific fishing techniques to yield varied output impacts (Huynh & Stringer, 2018; Malakar & Mishra, 2020). Hence, fishermen’s skills highly correlate with their human capital, influenced by factors such as education, training, and experience, as well as their physical and financial capital, family background, and environmental conditions. Additionally, each capital asset exists within a complex web of relationships with governance systems and policy institutions (Amir Zal, 2016; Bodin & Crona, 2008).
Individuals endowed with high adaptive capacity are less susceptible to external changes and, thus, better positioned to capitalise on opportunities for socioeconomic advancement (Lohmann, 2016). Sen’s influential entitlement approach ties vulnerability to inadequate access to assets, encompassing social capital; however, mere access to assets does not guarantee effective vulnerability reduction (Sen, 1983), as reflected in the SLF.
Various factors (multi-stressors) have led to fishermen’s vulnerability, particularly those involved in small-scale fisheries. It may be started with the degradation of common pool resources (natural capital). When one form of capital is weakened, it can negatively affect other forms. For instance, overfishing (depletion of natural capital) can reduce financial returns, weaken economic capital and limit fishermen’s ability to invest in better boats (physical capital). Similarly, suppose social capital is eroded due to conflicts or lack of trust. In that case, the cooperative ventures may fail, reducing the community’s ability to access the communal resources that benefit everyone. This vicious cycle further limits the function of each capital, making it harder to improve income levels, thereby perpetuating poverty. Most of the coastal fisheries (SSF) belonged to the lowest income group and have the following characteristics: These include: relying solely on one source of income, being skilled workers with low value-added, spending 62.6% of their income on basic amenities (food, housing, and transportation); facing increased living costs; having low levels of education and skills; being burdened with high levels of debt and being unable to own a house or property (Siwar et al., 2019). Since independence, various strategic programmes have been implemented to sustain this industry. Government initiatives in the form of capital, technologies, equipment, knowledge, and training have benefitted some fishermen who are actively involved in the commercial industry. However, the success rate remains low (Aisyah & Hayati, 2013). The SSF in Malaysia can be regarded as a marginalised and vulnerable group despite receiving numerous government assistance. Thus, specific attention needs to be considered to form a targeted intervention by examining the effectiveness of SSF’s capital utilisation. The interaction between the SLF capital types affects the resilience of SSF in the case of natural disasters or economic shocks, such as fluctuating fish prices, market demand, or global events. However, the complex and often fragile interconnections between these forms of capital are poorly understood, leading to a lack of comprehensive strategies to support their livelihoods.
SSF are among the most vulnerable communities in Malaysia, particularly in coastal states such as Kedah and Kelantan. They face numerous challenges, including environmental changes, economic instability, and limited resource access. As these challenges intensify due to climate change and resource depletion, it is essential to understand how SSF can build resilience and sustain their livelihoods. Despite the significance of these issues, limited research has focused on how various livelihood resources, such as skills, social networks, financial assets, and natural resources, influence fishermen’s ability to adapt. Many studies focus on various livelihood aspects of SSF; for example, climate change impact (Dzoga et al., 2018; Powell et al., 2022) and livelihood vulnerability (Mulyasari et al., 2020; Sadekin et al., 2021). Several studies also focused on the extent of adaptive capacity (Rubio et al., 2021; Silas et al., 2020) and drivers of adaptive capacity (Aguilera et al., 2015; Powell et al., 2022) of fishers. Multiple studies were conducted in various developing countries, including Ghana (Ameyaw et al., 2021; Antwi-Agyei et al., 2013), Indonesia (Mulyasari et al., 2020; Nissa et al., 2019), Bangladesh (Ahsan & Warner, 2014; Jakariya et al., 2020), and China (Chen et al., 2018, 2020), but few studies covered the fisheries community in Malaysia (Shaffril et al., 2017).
Hence, we aim to bridge this gap by proposing an integrated Adaptive Capacity Index (ACI) and exploring the interconnections among its major components. The index development process involves multiple steps: structural design, indicator selection, analysis scale determination, measurement error evaluation, data transformation, scaling, weighting, and aggregation (Robotham et al., 2019; Senapati & Gupta, 2017). Prior studies focused on broader environmental and economic consequences, with less regard for how the main livelihood capital elements interact to determine resilience (Bennett et al., 2016; D’agata et al., 2020). To address this deficiency, we aim to examine different aspects of adaptive capacity within livelihood capital elements involving an under-researched population, guided by three research questions: (1) What are the critical elements of SSF’s adaptive capability in Malaysia? (2) How are these elements combined to form SSF’s livelihood base? (3) Are these elements consistent across various geographical regions in Malaysia?
Kedah and Kelantan were chosen as primary research sites due to their distinctive vulnerability and contributions to Malaysian fisheries. Both locations present unique environmental, socio-economic and governance challenges that affect SSF - providing ideal settings to examine Malaysia’s diverse marine ecosystems’ adaptive capacities. We employ a novel analytical method by devising the ACI with Structural Equation Modelling (SEM). This technique allows an in-depth exploration of factors affecting adaption capacity, providing an effective framework for evaluating resilience among Malaysian SSF.
The paper is organised as follows: Section 2 defines SSF’s adaptive capacity. Section 3 introduces the analytical framework, while Section 4 outlines the methodology. Sections 5 and 6 present the study’s results and implications, respectively. The final section summarises the findings.
Conceptual Framework
We use the SLF and resilience theory to examine how SSF communities adapt to challenges and opportunities (DFID, 1999). This study incorporated all the forms of capital (human, natural, social, physical, and financial) as specified in the SLF. However, in previous studies, most researchers combined financial and economic capital as one context variable (Ding et al., 2018; Ehsan et al., 2019; Malakar & Mishra, 2020; Scoones Ian, 2001). This study separated the financial and economic capital as this relationship is complex, and both capitals serve different purposes in assessing a firm’s or household’s risk and stability. Financial capital typically refers to the funds available for investment and operational activities (Pomeroy et al., 2020; Tikadar et al., 2022), while economic capital is a risk-based measure that quantifies the capital required to cover potential losses (Bolder, 2022). Financial capital alone cannot capture systemic and idiosyncratic risks, while economic capital requires sophisticated models to assess (Gkartzios et al., 2022; Pomorina, 2021).
In this study, financial capital is defined as the monetary resources accessible for investment. Access to loans and subsidies can increase investment in fishing activities, thus improving income potential (Amadu et al., 2021; Firdaus et al., 2014). However, low repayment rates and inadequate financial institutions often hinder the effectiveness of government initiatives such as credit programmes, restricting their participation in productive activities. Although some studies referred to economic capital as physical capital (Allison & Ellis, 2001). For this study, economic capital encompasses broader material assets, including land and property, that can generate income (Bourdieu, 1986; Gkartzios et al., 2022). The interplay between financial and economic capital is significant; for instance, access to financial resources can enhance the use of economic assets, thereby improving overall productivity (Allison & Ellis, 2001; Gkartzios et al., 2022; Scoones Ian, 2001). Thus, understanding financial and economic capital dynamics is crucial for enhancing economic stability and autonomy. In conclusion, an integrated approach to measuring SSF adaptive capacity can assess economic capital, the foundation for long-term growth and resilience in rural communities. In contrast, financial capital is crucial for immediate operating requirements.
Based on previous studies, we analysed the direct and indirect links between these capitals using adaptive ability and livelihood capital (Antwi-Agyei et al., 2013; Huynh & Stringer, 2018; Robotham et al., 2019). We modelled capital interdependence to link them. Before this, we gathered relevant indicators from previous studies explicitly related to the SSF context. We selected these indicators using the statistical principal component analysis (PCA) result, which satisfied all validation requirements. These included a total variance explanation of more than 50%, a significant Bartlett’s test (
Defined Variables by Indicators.
The Roles of Physical, Natural, Social, Financial, and Human Capital in Shaping Economic Capital
Human, physical, and natural capital interaction drives economic success. Optimising physical capital use and production requires investing in human capital through knowledge, skills, and adaptive ability. Sustainable management and maintenance of natural capital ensure the long-term availability and resilience of fishery resources, which impacts SSF’s economic capital (Bennett et al., 2021). Human capital supports decision-making and adaptation; physical capital offers tools and infrastructure, and natural capital supplies SSF resources. SSF need financial resources to invest in their businesses, purchase essential supplies, and manage economic risks. Financial capital stimulates economic activity, income stability, and resistance to economic shocks (Senapati & Gupta, 2017; Shaffril et al., 2017). These varied capitals must be acknowledged as interdependent. Sustainable management and nurturing of human, physical, financial, social, and environmental capital is essential to optimise economic capital and sustain small-scale fishing communities.
The Roles of Social Capital in Human, Financial and Natural Capital
Strong social links enable knowledge sharing, information access, and skill development, boosting SSF’s human capital. Social capital, such as assertive community networks and shared norms, also facilitates natural resource protection collaboration (Bodin & Crona, 2008). Financial resources enable social capital cultivation and maintenance. Credit or microfinance allows SSF to invest in community-based organisations, cooperatives, or associations, developing social capital through collective action, pooled resources, and cooperation (Pomeroy et al., 2020). Conversely, natural capital depletion can weaken social cohesiveness and collaboration, emphasising the significance of conservation and sustainable management of natural resources, which require community cooperation (Allison et al., 2009).
Malaysia scored high in adaptive capacity relative to exposure and sensitivity. Behind assets, social organisation is Malaysia’s second most critical adaptability element (Hughes et al., 2012). Notably, a study by Malakar and Mishra (2020) examining recovery mechanisms after the Phailin cyclone found that linking and bonding social capital became significant only over time, not immediately or remotely following a disaster. While linking social capital correlates with recovery positively, bridging social capital, which is associated with conflicts over access to resources and aid, did not exhibit the same effect. Within the framework of this study, we hypothesise the following relationships:
The Relationship Between Human and Physical Capital
Productivity (Q) within the context of SSF’s livelihoods can be represented as a function of physical capital (K) and labour (L), as expressed by the equation Q =
Physical capital, exemplified by boats and equipment, is a learning and skill acquisition platform, as fishermen actively engage with these assets in their daily fishing operations (Färe et al., 2017; Prestrelo et al., 2019). Through hands-on experience, they develop specialised skills and techniques and deepen their understanding of the intricacies of their physical assets. This experiential learning process is evidence of the vital role played by human capital in adapting and enhancing physical capital.
Fishermen’s knowledge and skills empower them to make necessary modifications, perform repairs, and implement upgrades to their assets. In doing so, they not only extend the life and functionality of their equipment but also contribute to sustainability and the optimal utilisation of resources (Coglan & Pascoe, 2007; Viswanathan et al., 2001). This connection between human and physical capital emerges as a linchpin in enhancing the overall resilience and productivity of SSF. Thus, we hypothesise:
The Relationship Between Physical and Natural Capital
In fisheries, overcapitalisation occurs when the aggregate capital stock exceeds what is necessary for resource harvesting. This imbalance between fishing capacity and the available resource base has adverse ecological and economic consequences. Overcapitalisation can stem from various factors, including excessive investments in fishing technology, increased fishing efforts, government subsidies, or insufficient fisheries management measures (Chiat Lee & Viswanathan, 2019).
Such an imbalance can lead to overfishing, fish stock depletion, diminished fishermen’s profitability, and overall ecosystem degradation (Lohmann, 2016). Effectively managing overcapitalisation is imperative to ensure fisheries’ long-term sustainability and viability. This involves aligning fishing capacity with the productive capacity of the resource and implementing appropriate measures to control and regulate fishing activities (McClanahan et al., 2015). In this study, we propose the following hypothesis:
Building on the overview above, we have developed a study framework (Figure 2) incorporating selected indicators. All indicators are measured on an ordinal scale, ranging from low to high rankings. Within this structural equation model (SEM) study, we distinguish between two types of constructs: formative and reflective. Human, physical, economic, and financial capital are considered formative constructs, while natural and social capital are reflective constructs. These constructs are assessed based on a perception scale ranging from 1 to 5.
Methodology
Study Area and Data Collection
This study focused on SSF since they are one of the most disadvantaged B40 communities, representing the lowest 40% of Malaysian household income. Kedah and Kelantan were chosen for this study based on several key factors. Generally, both states have low economic status. Kedah, with RM 4,325, and Kelantan, with RM 3,563, often have the lowest household median incomes in Malaysia, resulting in the highest concentrations of households falling within the B40 group (earning less than RM 4,850). Kedah and Kelantan had the highest absolute poverty rates in 2018, at 8.8% and 12.4%, respectively. For comparison purposes, this study selected both states to indicate any differences in contribution factors to their adaptive capability in terms of geographical location, demographic background, authority management, and others.
Malaysian coastal fishing communities earn an average of RM 1,500.00 per month and operate under the Zone A fishing classification. Kedah has 9,471 licenced Zone fishermen with 4,109 vessels in 2019, while Kelantan has 2,684 fishermen and 1,200 vessels (Department of Fisheries Malaysia, 2019). We gathered data between January and September 2022 through stratified random sampling, specifically involving SSF registered with the Area Fishermen’s Association (PNK). Based on the information from the Kedah Fishermen Association (NEKAD) and the Kelantan Fishermen Association (PENEKA), researchers contacted the heads of the communities in identified areas to obtain details of the community members. We then conducted an initial field survey based on the average monthly income data to identify areas and houses. We assigned a serial number to each house as the first step in selecting a random number of sampling techniques. Subsequently, we selected the respondents using the results from the random number of samples generated by Microsoft Excel. Based on the selection, we distributed the questionnaires and recruited only the head of the household to participate in the survey.
A total of 730 Zone A fishermen were selected from nine fisheries areas (Figure 1) consisting of 398 SSF from Kedah and 332 SSF from Kelantan. However, after eliminating incomplete forms and applying sample corrections based on Mahalanobis distance, the final number of respondents (
Study area.
Structural Equation Modelling and Analysis Procedure
The selection of the PLS-SEM approach for this study is underpinned by its capacity to facilitate regression between formative and reflective constructs without requiring assumptions on variable distributions. In terms of the scale of measurement, PLS-SEM works with metric data and quasi-metric (ordinal) scaled variables and the standard PLS-SEM algorithm is able to accommodate binary-coded variables but requires additional considerations (Hair et al., 2021; C. Zhang et al., 2019). In addition, PLS-SEM is particularly appropriate in studies testing complex relationships, which involve both mediation and moderation effects (Mathies & Ngo, 2013; Mikalef et al., 2020; O’Cass et al., 2012). Hence, this quantitative approach has been widely applied in marginal group livelihood studies (Ibrahim et al., 2018; Longdong et al., 2020; Mathies & Ngo, 2013; Robotham et al., 2019; Yoon et al., 2015).
Distinct evaluations were conducted to assess the measurement model, as this study encompasses formative and reflective constructs, following the guidelines outlined by Hair et al. (2021). For formative constructs, the assessment criteria involved convergent validity, collinearity, and the statistical significance of indicator weights. In contrast, the measurement model for reflective constructs encompassed an analysis of individual indicator reliability, including outer loadings (λ), composite reliability (CR), and average variance extracted (AVE).
AVE quantifies the percentage of explained variance attributed to the variable. Acceptance criteria dictate that CR should exceed 0.6, and AVE should surpass 0.5 to meet the minimum standards. In assessing convergent validity, the discriminant validity rule dictates that the AVE’s square root for each variable should be greater than its highest correlation with any other variable. This criterion ensures that each variable captures unique phenomena not represented by other measured variables in the model.
Development of an Adaptive Capacity Index
Both Principal Component Analysis (PCA) and Partial Least Squares (PLS) are methods for calculating weights for variables in a composite index. PCA primarily serves to identify patterns within a dataset by reducing its dimensionality, effectively summarising the data’s variance. PCA is a standard method. However, it may fail when informative variations account for only minor variances in the variables (Yoon et al., 2015).
Conversely, PLS capitalises on the association between outcome variables and variables in a composite index, making it a viable alternative (Hair et al., 2021). The factor loading method focuses on elucidating the latent factors that underlie the observed patterns in the data. In terms of performance, a simulation study found that PLS is either as good or significantly better than PCA (Sihombing et al., 2023). Furthermore, PLS can handle non-metric variables frequently encountered in practice while building composite indices (Robotham et al., 2019). As a result, PLS is advised for practitioners who need to calculate weights for non-metric variables in composite indices (C. Zhang et al., 2019).
Selection of Indicators and Establishing Parameter Weights
Composite indicators are designed to capture a specific phenomenon or concept by considering multiple relevant factors or aspects (Robotham et al., 2019). Notably, in this study, only significant indicators are chosen. The weighting assigned to each parameter is derived from the outer loadings (λ) obtained through the output of path analysis. Additionally, this model accommodates interactions between variables. As a result, the weightage of the regression coefficient (β) in multiple regression also signifies the importance of the interactions within the variables rather than analysing them in isolation.
Calculation of the Index
The composite index is derived from the PLS-SEM analysis using the weighted sum scores approach. These weights are then used to calculate statistical parameters such as means, variance-covariance matrices, and PLS-discriminant components, which reveal product discrimination directions (Robotham et al., 2019). This method does not require the ordinal response to be converted to binary indicators, resulting in smaller weight matrices and avoiding the need to guess correlations of indicator variables for distinct categories (Nikoloulopoulos, 2015).
In this study approach, each respondent’s ACI is computed by combining factors extracted from the scores of various adaptive capacity indicators. The process begins by calculating an individual Cumulative Average Score (CSA) for each variable, which is then multiplied by the weightage determined from the outer loadings. Subsequently, to incorporate the influence of other variables that contribute to a specific variable, the index value for each variable is multiplied by the sum of regression coefficient values, denoted as β1, *β…, *βn, which signify the interconnectedness of these variables. This calculation also considers the roles of variables that emerge as mediators. The estimated scores (outer loadings), as presented in Figure 2 and the path coefficient values (β; see Appendix B) are employed as input for the calculation of the adaptive capacity index, following the formula:
Final model (PLS-SEM).
The total ACI is then calculated as the sum (∑) of each capital in the model. The final total ACI is standardised by using the min-max normalisation method shown in Equation 2 to obtain values lying between 0 and 1 (Hahn et al., 2009; Ibok et al., 2019).
We adopted a similar criterion in ACI and other level indices comparisons, categorising adaptive capacity value into percentiles as outlined in (Chen et al., 2020; Liu et al., 2013; Robotham et al., 2019).
Results
Measurement Model Evaluation
The formative construct has been validated, satisfying all the formative measurement model assessment criteria. No collinearity issues were observed, as indicated by the variance inflation factor (VIF) ranging between 1.019 and 3.824, all below 5. Figure 2 illustrates the outer loading values with inward arrows pointing to the latent variable representing formative constructs, while the outward arrows depict reflective constructs.
For the reflective constructs, specifically social and natural capital, most of the variables’ outer loadings (λ) exceeded the moderate threshold value of 0.5, except for S5. Nevertheless, this deviation is acceptable because the AVE and CR values fell within an acceptable range. The CR values for all capitals exceeded 0.6 (natural capital CR = 0.781 and social capital CR = 0.845), indicating a high internal consistency reliability within the model. Convergent validity was successfully demonstrated by effectively measuring the related variables, as indicated by the AVE values. This is further supported by the evidence of discriminant validity, where the squared AVE values for both the natural capital (0.797) and social capital (0.702) constructs surpassed the correlation between the constructs (.241). These findings suggest that the indicators measure distinct concepts of latent variables.
Structural Model Evaluation
The path coefficients of the standardised regression measure the strengths of the relationship between the capitals or the proposal of some hypothesis of causality. The t-values (see Appendix B) show the standardised regression coefficients (β), which are significant at 5%. The physical and natural capital influenced the economic capital but not the human capital. The human capital was also unable to influence the financial and physical capital. Social capital significantly mediated the relationship between physical and economic capital and natural and economic capital. Increased social capital is associated with increased physical and natural capital, ultimately influencing economic capital. Financial and physical capital interaction shows a borderline significant relationship with economic capital. It suggests that the joint effect of financial and physical capital may positively impact economic capital.
Having confirmed that variable measurements are reliable and valid, the structural model is evaluated, examining the relationship between the variables and the model’s predictive capabilities. The assessment procedures include
As outlined in the ACI calculation from the methodology section, Table 2 shows the cut-off value for each quartile in every index component, ranking them from very low (<25%), and low (26%–50%), to moderate (51%–75%) and high indices (>76%).
Adaptive Capacity and SLF Capitals Percentile Range Indices.
Discussion
The Analysis of Income Thresholds
The analysis of income thresholds and adaptive ability provides valuable insights (see Table 3). The income range between the “B1 to B2” thresholds is explicitly associated with extremely low and poor adaptive capacity cases. This relationship raises questions about the challenges this income class faces, significantly impeding their ability to adapt to changing circumstances. On the other hand, the B4 income threshold is correlated with high adaptive capacity. These findings underscore the impact of income on adaptive capacity. The prevalence of very low and low adaptive capability within the B1 to B2 income threshold; emphasises the need for targeted interventions and support systems to assist financially strained individuals and households in adapting and mitigating vulnerabilities (Abu Samah et al., 2019; Senevi Gunaratne et al., 2023).
ACI Level by Income Threshold.
However, even though the B1 income threshold was identified as the most affected group, some could still develop their capacity at a moderate and high level. According to the contingency table analysis (See Table 4), the chi-square test statistic revealed that variables such as age, years of working, number of household equipment, vessel age, length, and engine horsepower were associated with their adaptive capacity level.
Cross Tabulation Analysis B1 Income Threshold versus Selected Indicators.
Identification of Capitals and ACI Across Districts
In analysing each capital index across districts (See Figures 3 and 4), most districts reported a low economic capital index, signifying their low capability to translate that capital into income generation and increase the standard of living. While lowest recorded in Semerak, Kuala Muda’s location near the high-income state of Pulau Pinang contributed highest to its economic capital index score but fares less in other categories, specifically in human and physical capital, thus resulting in their weak score in adaptive capacity involved more than 50% of them. The strong social cohesion is evident among the SSF in Tanjung Dawai, as indicated by their high ranking on the social capital index. The strategic positioning of Tanjung Dawai, situated in the estuary, has facilitated the SSF in effectively managing the influx of external entities while developing vital interconnections among themselves to safeguard their fishing grounds. Communities in Tanjung Dawai are also actively engaged in entrepreneurial activities, as this place is known for its seafood and dried marine products. This highlights the significant role of social capital in facilitating resource access, building partnerships, and creating collaborative opportunities in business settings as Bodin and Crona (2008) suggested. Furthermore, the fishermen’s testimonies from Langkawi during our survey have substantiated their firm association. Nevertheless, despite their strong social capital, Tanjung Dawai’s lowest scores in financial and physical capital signify some obstacles in obtaining and effectively utilising such capital, explaining their highest numbers in the very low adaptive capacity group.
SLF capitals index by districts.
Percentage distribution of ACI level by districts.
In the natural indices analysis, all Kelantan districts exhibit lower values compared to all districts in Kedah. Kelantan’s geographical location along the northern part of the eastern coast of the Malaysian Peninsula exposes its four coastline areas to the South China Sea. There are no islands or boundaries beyond the sea line to shield the coastline from natural processes driven by the ocean system. Despite the impact of development activities, Kelantan’s coastline continually changes; this index result aligns with Wahid and Osman (2023) findings, which studied changes in ocean wall direction and their effects on sea wall morphology. While some villages along the shoreline in Tumpat experience erosion, other seaside areas remain vulnerable to degradation. Between 2000 and 2019, significant changes occurred along the coastline, with the degradation rate persisting despite ongoing aggregation efforts (Azali, 2022). Notably, northern areas like Kampung Pauh Seratus, Kampung Pantai Geting, and Kampung Sri Tujuh continue to advance towards the sea. The natural sources in all districts in Kelantan are so critical that they potentially contributed to their significant numbers in very low and low ACI, ultimately in the areas of Tumpat and Semerak. However, despite natural risks, we identified high physical capital indices among SSF in Kota Bharu. This may also be used to characterise the inventiveness of eastern coast residents, who are skilled and well-known for their shipbuilding expertise. The low and moderate levels of ACI dominated the SSF households in Kota Bharu. On the other hand, Langkawi in Kedah, known for being an island surrounded by 99 islands, has boasted the highest adaptive capacity through its highest natural index, indicating the success of its sustainability programme. Since 2006, over 224 artificial reefs have been installed, contributing to the region’s productive conservation efforts (Department of Fisheries Malaysia, 2019).
Langkawi also recorded the highest level of human capital, as indicated by their receipt of additional training beyond fishing skills. This might be attributed to Langkawi being a popular tourist destination. Therefore, a significant number of fishermen in Langkawi may acquire knowledge and expertise by engaging in supporting services within the tourism and hospitality industry. For instance, many individuals can communicate in English and possess the skills to facilitate various tourism-related activities. Overall, Langkawi performs well in most categories, except for the financial capital index.
The high financial index in Bachok and Yan has potentially explained the effectiveness of financial capital utilisation as both districts recorded high ACI. Despite an average score in every capital category, SSF in Bachok increased their ACI. This district’s local economy predominantly relies on agriculture, with significant fishing activity. Recently, Bachok Town was designated as an Islamic Tourism Town due to its beautiful beaches, such as Pantai Irama, Pantai Kandis, Pantai Kemayang, Pantai Melawi, and Pantai Pengkalan Pateh (Bachok District Council, 2016). Moreover, numerous tertiary education institutions in the vicinity suggest the potential for future economic growth, justifying the relatively high financial capital score. As for Yan, this district has benefitted from numerous new federal and state projects that catalysed more economic activities, that is, the Pulau Bunting in Yan, which has been designated as a power plant and gas supply centre and a tourism destination.
This study then investigates how the interconnection of the capitals contributes to differentiating the level of each capital index as summarised in Figure 5. Financial and physical capital exhibited the most significant positive differences between low and high human capital index communities, implying that having more financial capital means being able to purchase more advanced fishing gear or boats (physical capital), which is critical for human capital development because it provides necessary assets and equipment for the SSF learning process and skills development (Freduah et al., 2017; Pomeroy et al., 2020). In addition, increasing financial access may allow them to allocate more resources to education and healthcare (Ahsan & Warner, 2014; Hahn et al., 2009). Individuals in the high human capital group are becoming more self-reliant and pursuing more individualistic actions, leading to a reduction in social capital and a decrease in reliance on their community networks as suggested by Coglan and Pascoe (2007), a thriving fishing community is inextricably linked to family history. Meanwhile, equal resource utilisation between high and low groups demonstrated that human capital development was not supported by using natural and economic capital; rather, natural and economic capital are highly optimised when high-quality human capital exists, as mentioned in the sequence section.
Web analysis of capital utilisation effectiveness.
Societies with more economic capital exhibit significantly higher levels of natural and physical capital, indicating improved resource management and effective management of their physical capital, which leads to an increase in their economic capital. This link indicates that wealthier societies can reinvest their resources in natural and physical assets (Färe et al., 2017). Societies with high economic capital have slightly lower financial capital, suggesting that financial resources may be less important when other forms of capital, mainly natural and physical capital, are highly developed (H. Zhang et al., 2020). Conversely, low economic capital groups have higher financial capital, suggesting they may rely on financial resources through loans, savings, or subsidies (Chen et al., 2020; Chiat Lee & Viswanathan, 2019). Social capital also rises slightly in larger economic capital communities, possibly reflecting more stable and resilient social structures, which may help foster collaboration and support systems (Antwi-Agyei et al., 2013; Robotham et al., 2019).
High physical capital groups have higher financial capital, which makes sense given that financial resources are frequently necessary to invest in physical infrastructure (e.g., homes, fishing boats, farming equipment, and roads). The high economic capital also has shown wealthier communities have better access to financial resources, allowing for more investment that can increase productivity and living conditions (D’agata et al., 2020; Prestrelo et al., 2019). Communities with little financial capital may struggle to purchase or maintain such assets, resulting in lower physical capital levels. Social capital is higher in low physical capital societies, implying that in the absence of physical assets, this group relies more on strong social networks and collaboration (Bodin & Crona, 2008). However, in high-capital societies, individuals rely less on social bonds because they can meet their needs with improved physical assets and financial resources (Amir Zal, 2016; Cordaro et al., 2021).
Communities with high natural capital show a higher human capital score, suggesting that access to or abundant natural resources (such as fisheries, forests, or fertile land) is positively influenced by greater skill and working experience. Interestingly, financial capital is higher in communities with low natural capital, suggesting that when natural resources are limited, communities might compensate by developing stronger financial resources that are not reliant on natural capital. This may reflect a shift towards financial strategies such as loans, savings, and investments to sustain livelihoods in resource-poor environments (Huynh & Stringer, 2018; M. M. Islam et al., 2014). In contrast, communities with high natural capital might rely more directly on their natural resources for income generation and may therefore have less need to build up financial reserves (Cordaro et al., 2021).
Human capital, financial and physical capital are higher in low social capital communities which is consistent with the previous finding that this group rely less on social capital (Amir Zal, 2016; H. Zhang et al., 2020). This comparison reinforces the idea that social capital can provide a cushion for marginal communities that rely highly on social capital, allowing them to distribute resources and reduce individual burdens (Ahsan & Warner, 2014; Huynh & Stringer, 2018), while communities with weaker social ties may need to rely more on personal resources (Barnes-Mauthe et al., 2015).
Human capital is higher in high financial capital communities, suggesting that these communities are better able to leverage or accumulate financial resources when human capital is high. As financial capital is vital for physical capital access, financial capital also benefits from better physical infrastructure, as infrastructure investments often lead to economic growth and increased access to financial resources (Färe et al., 2017; Pomeroy et al., 2020). Economic capital sees a slight increase in high financial capital communities, indicating a minor positive correlation between financial resources and overall economic prosperity. On the other hand, natural capital is higher in low financial capital communities, suggesting a greater reliance on natural resources in less financially secure environments, while wealthier communities may diversify into other capitals (Prestrelo et al., 2019). Social capital is marginally higher in low financial capital communities, indicating stronger community ties where financial resources are scarcer (Antwi-Agyei et al., 2013; Shaffril et al., 2017).
Overall, social capital plays a crucial role in the low financial and physical capital communities, and SSF depends on financial resources in the form of loans, savings, or subsidies to compensate when economic and natural capital is scarce. On the other hand, the low financial capital group will depend on natural resources and social network support to make their way. This dynamic underscores how different forms of capital act as complementary or compensatory mechanisms depending on a community’s resource constraints (Tinch et al., 2015; H. Zhang et al., 2020). This result also highlighted the ability of the group with strong financial and physical capital to invest in the development of human capital while also leveraging social networks and natural resources to boost their economic capital (Amadu et al., 2021). Our findings illustrate the complex environment of adaptive capacity among SSF in Malaysia, which depends on the interplay among the main capitals.
Theoretical Implications
The study contributes to the theoretical understanding of geographically distributed adaptive capacity by highlighting the importance of considering location-specific factors in vulnerability assessments and adaptation planning. By assessing various capital indices (e.g., financial, natural, human, social, economic, and physical) to evaluate adaptive capacity, the study enriches the theoretical framework for understanding the interplay between different forms of capital in shaping adaptive capacity. By exploring interactions among various forms of capital that affect resilience in Kedah and Kelantan, our study adds an essential perspective to discussions of resilience versus vulnerability within SSF. Additionally, our study contributes to the resilience literature by revealing specific regional differences in the adaptive capacities of Kedah and Kelantan, emphasising the significance of socioeconomic conditions and environmental vulnerability and underlining targeted adaptation strategies commonly referenced in resilience studies but unexplored within Malaysian fisheries. By examining these capitals in two different geographical locations, the east coast and west coast of Peninsular Malaysia, the study provides theoretical insights into how location influences adaptive capacity, often overlooked in broader national- and regional-level analyses. This emphasises the need to comprehensively understand local contexts in developing adaptation strategies.
Policy Implications
Our findings offer several practical implications for policymakers, aiming to enhance adaptive capacity and resilience to multidimensional vulnerabilities for SSF in Kedah and Kelantan. These implications include:
First, prioritise targeted capacity-building programmes. In districts with low adaptive capacity, such as Kuala Muda and Tanjung Dawai, local governments should focus on capacity-building programmes aimed at enhancing technical and management skills, especially in physical asset acquisition. These programmes could be organised in partnership with local universities, NGOs, and fisheries organisations to ensure they meet the specific needs of SSF (Radin Firdaus et al., 2020). By providing training in aquaculture, sustainable fisheries practices, and financial management, these programmes can increase the economic resilience of fishermen.
Second, invest in infrastructure development for resilience. In districts with weak infrastructure, local governments should upgrade fishing equipment, jetties, and other essential infrastructure. This can be done by applying for federal or regional funding dedicated to coastal development. Partnerships between local municipalities and private sector stakeholders can accelerate infrastructure improvements, directly enhancing physical and economic capital for SSF.
Third, promoting economic diversification. Local governments can encourage SSF to diversify their income sources by investing in aquaculture projects. This could be achieved by providing grants, low-interest loans, or subsidies to fishermen willing to explore aquaculture alongside traditional fishing practices. Organisations like the Fisheries Development Authority of Malaysia (LKIM) could work closely with SSF to provide guidance and resources for such transitions. Investments in tourism and education, as in Bachok and Yan, can enhance financial capital and resilience.
Fourth, implementing coastal protection initiatives. Given the Kelantan districts’ low natural capital index and ongoing coastal changes, policy efforts should prioritise coastal protection measures. Lessons from Tumpat’s experience, as highlighted in Azali (2022), can guide effective coastal management strategies. As seen in Langkawi, local governments can collaborate with environmental organisations to implement mangrove restoration, artificial reef projects, and sustainable coastal management practices. These actions will preserve natural capital and support long-term fisheries sustainability.
Finally, strengthening community engagement and networking. Policymakers should promote community engagement and collaborative partnerships to strengthen social capital. In partnership with NGOs, local government offices can organise community forums and networking events to foster collaboration among SSF, share best practices, and create cooperative structures. These initiatives can leverage the successful example of Tanjung Dawai, where strong social networks have bolstered economic activities.
Conclusion
This study offers valuable insights into the adaptive capacity of SSF in Malaysia, using the ACI within the SLF. The findings reveal the critical role of financial and physical capital in shaping economic resilience, while social capital enhances the management of natural resources. The differences in adaptive capacity across regions highlight the importance of developing location-specific adaptation strategies. Specifically, Semerak and Tumpat require concentrated capacity-building and infrastructure improvements to prevent future deterioration of natural capital. In contrast, financial and physical capital investments in Kuala Muda and Tanjung Dawai will enhance the SSF ACI. Moreover, the results suggest that promoting aquaculture and sustainable fishing practices can enhance economic stability and alleviate pressure on marine resources. These findings could inform policymakers in designing interventions that effectively enhance the resilience of SSF.
Limitations and Future Research Directions
Considering the current study’s limitations, future research should involve a larger sample size (more than 750 respondents) to validate the ACI. Furthermore, since the ACI validation only encompassed two states in Malaysia, it is necessary to revalidate the indicators for other states. While this study is of immense value, specific issues require consideration. First, using cross-sectional data can make it harder to appreciate how adaptive capacity changes over time and may obscure any associated resilience strategies or external stressors. Furthermore, while the ACI provides an inclusive measure of adaptive capacity, its weighting may not accurately represent each fishing community’s specific circumstances; as a result, adjustments must be made according to contexts. Furthermore, its focus on Kedah and Kelantan may provide greater insight into these regions’ economies while potentially restricting generalisation to Malaysia’s coastal regions.
Future research should address its limitations and deepen our understanding of SSF adaptive capabilities. Studies performed over time could clarify how adaptive capacity changes due to changes in environmental circumstances, policy interventions and market forces. Enhancing and validating the ACI across different contexts could enhance its usefulness to practitioners and policymakers looking to strengthen fisher communities’ resilience. Participation in research that directly engages fishermen could ensure future strategies are practical and suitable to their culture, that is, by studying the effects of globalisation, digitalisation, and climate change on SSF adaptation strategies.
Footnotes
ORCID iDs
