Assessment of the Vulnerability of Households Led by Men and Women to the Impacts of Climate-Related Natural Disasters in the Coastal Areas of Myanmar and Vietnam

Abstract

Farm households along the coastlines of Myanmar and Vietnam are becoming increasingly vulnerable to flooding, saltwater intrusion, and rising sea levels. There is little information available on the relative vulnerability of men- and women-headed households, and the governments of Myanmar and Vietnam have not identified or implemented any adaptive measures aimed specifically at vulnerable peoples. This study aims to fill these gaps and assess the relative climate change vulnerability of men- and women-headed farm households. This study considers 599 farm households from two regions of Myanmar and 300 households from Thua Thien Hue province of Vietnam for the period 2021–2022. We offer a livelihood vulnerability index (LVI) analysis of men- and women-headed farm households using 46 indicators arranged into seven major components. The aggregate LVI scores indicate that farm households in Myanmar are more vulnerable (scores of 0.459 for men and 0.476 for women) to climate-related natural disasters than farm households in Vietnam (scores of 0.288 for men and 0.292 for women), regardless of the gender of the head of household. Total vulnerability indexing scores indicate that women-headed households are more vulnerable than men-headed households in both countries. Poor adaptive capacity and highly sensitive LVI dimensional scores explain the greater vulnerability of women-headed farm households. The findings also highlight the importance of the adaptive capacity components reflected in the LVI analysis in reducing farm households’ vulnerability.

1. Introduction

Climate change has exacerbated the rise of sea levels and poses an existential threat to some low-lying coastal areas, affecting the socioeconomic vulnerability of communities in coastline areas [1,2,3]. Sea-level rise, floods, and saltwater intrusion are becoming more common in the Mekong region and wreaking havoc on farming communities, livelihoods, and food and nutrition security. Nevertheless, little is known about the vulnerability to climate change in farming households in the coastal Mekong region, where there are scarce resources for creating and implementing adaptation strategies [4,5]. National adaptation strategies have focused on hard strategies such as infrastructure rebuilding; few have focused on identifying the vulnerability of various groups and communities and how climate adaptation funds might be better allocated to meet their needs. An assessment of the vulnerability of the communities and regions in the Mekong region has been particularly encouraged by development agencies and public and private organisations [4,6,7].

The countries in the Mekong region are extremely exposed to the effects of extreme climate events such as rising temperatures, erratic rainfall, flooding, and sea-level rise [4,5,6,7]. The Global Climate Risk Index report ranks Myanmar as the second most climate-hazardous country in the world [2]. The coastline areas of Myanmar lie just 3 m above sea level and are densely populated. Therefore, these areas are vulnerable to flooding, saltwater intrusion, and other adverse effects of climate change [6,8]. Since the main source of income in the delta areas is agriculture, the adverse impacts of climate change and the rise of sea level will put an additional burden on the people in these areas. Recent data show that rice farming in coastline areas is frequently impacted by saltwater intrusion and coastal flooding along with rising sea levels [9,10,11]. Vietnam is similarly vulnerable; it is currently sixth among countries that have suffered the most from climate change over the last two decades [2]. It is one of nine countries in which at least 50 million people are at risk of exposure to, among other things, increasing sea levels, saltwater intrusion, and powerful storms [12,13]. With over 3200 km long of coastline, shrimp aquaculture, rice-based cropping, and fisheries are three distinct systems in Vietnam, and these sectors are particularly vulnerable to adverse effects of climate change [14,15]. We have thus selected Vietnam and Myanmar as countries in the Mekong region with a high climate risk to assess the varying effects of farm households’ vulnerability to flooding and saltwater intrusion.

The effects of climate change are not evenly distributed, and marginalised groups, including women, are disproportionately affected. Female farmers frequently face gender-specific barriers to resources and decision-making power, exacerbating their vulnerability to climate risks [16,17]. In many regions, women play an important role in food production, processing, and marketing, and their vulnerability has consequences for agricultural productivity and food security. Understanding the gendered differences in the impacts of climate change is therefore crucial for developing effective and gender-sensitive adaptation strategies that address the specific needs and challenges of women farmers [16,17,18]. Women constitute the majority of the world’s poor and rely on natural resources that are most vulnerable to climate change [19]. There are nevertheless few research assessments of the relative level of climate change vulnerability experienced by women- and men-headed households where the causes of women’s and men’s vulnerability to climate change are many and varied [19,20,21].

As the United Nations Framework Convention on Climate Change (UNFCCC) [19] points out, simply adding women to existing enquiry frameworks and approaches is insufficient. Rather, the root causes of vulnerability need to be addressed by using a more nuanced approach to capture multiple and intersecting social pressures related to gender. Therefore, we conceptualise the root causes of vulnerability in men- and women-headed households by identifying multiple and intersecting indicators, social pressures, and climate-related natural hazards. The remainder of the study is structured as follows. In Section 2, we present theoretical conceptualisation, followed by our research methodology in Section 3, research findings and discussion in Section 4, and our conclusions and recommendations are presented in Section 5.

2. Theoretical Conceptualisation

The literature highlights that although climate change is a global process, there is a need to test climate change vulnerability at the community level [6,22,23]. Therefore, many scholars advise conducting localised assessments of farm households’ vulnerability and evaluating the vulnerability of communities and regions [10,23,24]. There are many methods available to assess the vulnerability and impacts of climate change on specific communities or societies; the most common are econometrics and indicator-based approaches [6,23]. Indicator methods are easy to use, and indications can be easily compiled into an index to measure various vulnerabilities [10,23,24].

Climate change impacts male and female farmers differently, depending on their social, economic, and cultural contexts, and it is important to identify and address the differential vulnerabilities of men and women farmers [18,19,25,26]. Aryal et al. [16] find that even when exposed to the same climate risks, households in Bangladesh’s coastal region face different levels of livelihood impact depending on whether they are led by men or women; these disparate livelihood impacts are largely explained by differences in access to assets and the availability of adaptation options. Thus, this study considers indicators reported across the scientific literature to assess the dimensions and contexts of climate-related vulnerability of women- and men-headed households in the coastline areas of Myanmar and Vietnam.

Since the combined impacts of climate change and water insecurity may be severe for communities in remote locations with limited infrastructure, we consider access to market and infrastructure accessibility to estimate the adverse effects of climate change on communities with or without limited infrastructure [27]. This may capture the effect of public investment in infrastructure as climate resilience strategies in Myanmar and Vietnam. Time spent travelling to source water is a well-known determinant of quality of life and determines the overall burden of fetching water [6,10,28]. Women often play a greater role than men in managing water supplies as they are the primary users of water in the household and are more likely to detect water insecurities. Women are also more burdened than men by household caregiving, including having larger roles in household food preparation [6,29,30]. This study explicitly considers some norms and cultural indicators related to fetching water, time taken for such collection, and caregiving of family members to better estimate the differences in climate change vulnerability of farm households in the coastline areas of Myanmar and Vietnam.

Some studies divide livelihood vulnerability to climate change into five major components based on five livelihood capitals identified in sustainable livelihood framework analysis. However, this would not as accurately reflect the intricacies of household vulnerability as would the seven major components of livelihood analysis that we apply here [6,10,31,32]. The seven major components of livelihood analysis are the following: socio-demographics, social networks, livelihood strategy, food, health, water, and natural hazards. According to the Intergovernmental Panel on Climate Change (IPCC)-defined vulnerability standard, the first three components can be constituted in the adaptive capacity dimension, while health, food, and water components can be constituted in the sensitivity dimension. The exposure dimension is used to characterise the natural hazard component [1,3].

This study introduces new indicators based on the outcomes of focus group discussions. These indicators reflect support from private, community-based organisations and public organisations during the COVID-19 pandemic. This study also considers new indicators related to the conflict between water users and the effect of this conflict on local communities in water resource utilisation and management at the farm level in Myanmar and Vietnam. The study assesses the difference in vulnerability of women and men households by indexing livelihood.

We adopt the indexing method applied in earlier studies [6,33,34,35]. However, our innovative contribution is that we employ indicators reported across various studies as well as local-specific and community-validated indicators. Our study offers concrete findings and has implications for policymakers and development planners in Myanmar and Vietnam. Second, using the 46 indicators, we apply the livelihood vulnerability index (LVI) and a gender-treatment effect to examine the climate change vulnerability of men- and women-led households in two countries.

3. Research Methodology

3.1. Study Areas and Data

This research is conducted in Vietnam and Myanmar. Due to their vast geographies and their populations’ limited knowledge and understanding regarding the influence of climate change, this study is focused on two regions of Myanmar and one region of Vietnam. A randomised sampling method is considered because of the complications and impacts of the COVID-19 pandemic in Myanmar and Vietnam; samples are mostly from areas in which travel was less restricted. For instance, Myanmar had fewer travel restrictions within communities at the time of the survey, but there were strict travel restrictions throughout Vietnam due to the COVID-19 pandemic. Initially, villages are selected based on particular characteristics, such as villages located near coastline areas and in which both agriculture and aquaculture farmers are located. Data are collected in four distinct steps: (i) field observation, (ii) validation of selected indicators with key stakeholders through focus group discussions, (iii) final questionnaire preparation, and (iv) farm household surveys or interviews conducted from September to October 2021 in Myanmar and from February to March 2022 in Vietnam (see Figure 1).

Taking into account the factors noted above, we select 12 sampling communities in Myanmar and six sampling communities in Vietnam along the coastline and delta areas. We select a sample of 300 individuals in each region for interviews. In Myanmar, our final sample, from which some observations were missing, comprises 599 respondents, and that for Vietnam is 300. Despite inconsistencies in the available dataset, the LVI method that we apply in this study uses a balanced weighted average approach to minimise the variance. To ensure the indicators selected are appropriate, these are tested and validated with the key stakeholders during the focus group discussions. Nineteen indicators are selected to represent the components of socio-demographic factors, livelihood strategy, and social network. Twenty indicators are considered for inclusion in the sensitivity dimensions (i.e., health, food, and water components), and seven indicators are selected for inclusion in the exposure dimension. The LVI analysis includes a final total of 46 indicators divided into seven major components.

In Myanmar, we selected six communities in Dedaye and Kyaiklatt Township in the Ayeyarwaddy region (Ka Lar Su, Kyon Thin, Hmyaw Sin, Kunpa Laing, Kyee Chaung, and Hpoe Shan Gyi villages), and six communities in Kyauktan Township in the Yangon region (Kamarmat, Oke Pho, Thet Kel Kone, Yae Kyaw, Shwe Pyi Thit, and Zwe Bar villages). The three townships that were chosen are all low-lying areas with elevations under 5 m; see Figure 2.

In Vietnam, we select six communities in three districts of the Tam Giang–Cai Hai lagoon: Quang Dien (Quang Loi and Quang Thai), Phu Vang (Vinh Xuan and Vinh Phu), and Phu Loc (Vinh Hung and Vinh Giang); see Figure 3. For the village profiles in the sampling areas, we developed gender- and ethnicity-related questions and worked to ensure the participation of women and minority groups in the research assessment.

Enumerators from the selected villages are chosen and trained through virtual training and pilot surveys; this approach is taken because of the impacts of travel restrictions during the COVID-19 pandemic and the political situation in Myanmar. The data collection tool, KOBO, is used to collect data and check the quality of the dataset. In Vietnam, the enumerator training is conducted at Hue University, and field-based data collection is carried out. All potential interviewees were asked to sign an ethical consent form to participate in the survey; interviews are only conducted with those willing to participate. The research team paid attention to the norms, values, and ethics of community participants and considered their ethical principles, such as following COVID-19 coping measures, keeping participant data anonymous, and practising secure data management during the research assessment, despite the development of COVID-19 pandemic in the region.

Although our study is in line with the proposed research objectives, it has certain limitations. The enumerators were trained and tested in advance of the survey using a set of questionnaires. Their level of understanding of the concepts of climate change and the types of questionnaires may not be perfect. This study may also be associated with the specificities of each of two different contexts or countries in one single questionnaire. However, we confirm that our data are of high quality through a daily data quality check. Moreover, this study focuses on the impacts of climate change on farm households and does not take into account the impacts on the ecosystem, biodiversity, livestock, and animals, and so on.

3.2. Development of Livelihood Vulnerability Index

Although climate change is happening globally, vulnerability is time- and location-specific. Deressa [23] and Tun Oo [6] recommend comparing and testing climate change vulnerability at the community level. However, others argue that the concepts used to understand vulnerability remain vague and are inconsistently defined [36,37], leading to different indexes. Others state that, in principle, vulnerability cannot be measured [38]. Hinkel [39] nevertheless presents four arguments for developing vulnerability indicators. The indicator approach has weaknesses and strengths; however, due to the simplicity of constructing an index, it is applied in many types of research related to climate change vulnerability assessment [39].

In this study, we selected and validated indicators to convey farm household vulnerability to flooding and saltwater intrusion in both countries (see detail in Appendix A). Each component comprises several indicators, and these are standardised using a balanced average approach [10,22]. The major component index of the men- and women-headed households can be calculated after each index has been standardised following Equation (1):

$$M_h=\frac{{\sum }_{i=1}^{n}index{S}_{h}i}{n}$$

(1)

where $M_h$ denotes one of the seven major components of men- or women-headed households ‘h’; $\mathit{i}\mathit{n}\mathit{d}\mathit{e}\mathit{x}{\mathit{S}}_{\mathit{h}}$ represents the subcomponent index or indicator denoted by ‘$i$’ of households ‘$h$’ and ‘$n$’ the number of indicators applied in each major component. After the major component of men- and women-headed households ‘$h$’ for each of the seven major components is deduced, the LVI of men- and women-headed households is calculated as follows in Equation (2):

$${LVI}_h=\frac{{\sum }_{i=1}^n{M}_h}{n},$$

(2)

where ${LVI}_h$ is the livelihood vulnerability indexing scores for the man- or woman-led household ‘$h$’; ‘$M_h$’ represents one of the seven major components by household type ‘$h,$’ ‘$i$’ is the index of farm households; and ‘$n$’ is the number of major components. As the balanced average approach was applied, the LVI was aggregated with the index derived from the average scale of each indicator. The contribution factor such as exposure, sensitivity, and adaptive capacity scores can be calculated as the following Equation (3):

$${CF}_h=\frac{{\sum }_{i=1}^n{M}_{ih}}{n}$$

(3)

where ${CF}_h$ is an IPCC-defined contribution factor for households ‘$h’$, $M_{ih}$ are the major components for households ‘$h’,$ and ‘$n’$ is the number of major components in each contributing factor. Once the exposure, sensitivity, and adaptive capacity dimensions of vulnerability are calculated, the IPCC-defined livelihood vulnerability index ($\mathrm{IPCC}\_\mathrm{LVI}$) can be calculated as the following Equation (4):

$$\mathrm{I}\mathrm{P}\mathrm{C}\mathrm{C}\_\mathrm{L}\mathrm{V}\mathrm{I}=\left(e_h-a_h\right)\times s_h$$

(4)

where $e_h$denotes the exposure dimension of the farm households ‘$h’$, $a_h$ is the adaptive capacity of farm household $‘h’,$ and $s_h$ is the sensitivity dimension of the farm households ‘$h’$. According to the IPCC [1], adaptive capacity is the ability of a system to adjust or respond to climate change impacts and cope with consequences. Climate change exposure refers to the extent to which a system is subjected to climate change impacts such as floods, drought, heat waves, and sea-level rise. Climate change sensitivity refers to the degree to which a system is affected by climate-related factors [6,10,40].

4. Results and Discussion

In this study, out of 899 respondents, 603 men-headed households and 296 women-headed households participated in the survey during the survey periods in 2021 and 2022. Women-headed households are defined as farm households currently under the supervision of women in the absence of (or as a result of the death of) a male household head and with farm activities mostly undertaken by the women in the household. While we intended to include an equal number of men- and women-headed households in the survey, only around one-third of women-headed farmers could be included. As a result, only 222 (37.1%) women farmers participated (out of 599) in Myanmar, and 74 (24.67%) women farmers participated out of 300 respondents in Vietnam. In terms of ethnicity, 509 respondents (85%) are of the Bamar ethnic group, and the remaining respondents are of other ethnicities, including 41 Hindu (6.8%) and 43 Kayin (7.2%) farmers in Myanmar.

4.1. Component-Based Vulnerability: LVI

The aggregate LVI score is the distributed output of the seven major components for a total of 46 indicators.

Table 1. Calculated major components and subcomponent index scores of LVI for men- and women-headed households in Myanmar and Vietnam.

Major Component	Indicators	Unit	Men-VN (226)	Women-VN (74)	Men-MM (377)	Women-MM (222)
Socio-demographic	Average Age (years)	Year	0.548	0.411	0.445	0.432
	A household without electricity (0: yes; 1: no)	%	0.013	0.092	0.491	0.586
	Households without secondary education (0: yes; 1: no)	%	0.237	0.303	0.485	0.631
	Without work for family men (0: yes; 1: no)	%	0.158	0.25	0.013	0.135
	Without work for family women (0: yes; 1: no)	%	0.171	0.276	0.223	0.135
	HH with family members less than 15 years (dependent children=number)	%	0.263	0.237	0.585	0.61
	Balanced Average Score (St.dev)		0.232 (0.03)	0.262 (0.017)	0.374 (0.036)	0.422 (0.039)
Livelihood strategies	Household agriculture as the main income source (1: yes; 0: no)	%	0.408	0.513	0.393	0.356
	Household aquaculture as the main income source (1: yes; 0: no)	%	0.355	0.224	0.255	0.217
	Households without having income jobs at the moment (0: yes; 1: no)	%	0.526	0.474	0.549	0.559
	Households without income from non-farm sources (0: yes; 1: no)	%	0.526	0.421	0.533	0.541
	Households without incomes from remittance (0: yes; 1: no)	%	0.763	0.303	0.822	0.788
	Households without income from off-farm sources (0: yes; 1: no)	%	0.5	0.724	0.714	0.73
	Households difficult to repay loans (1: yes; 0: no)	%	0.355	0.421	0.676	0.73
	Household currently without access to any credit (0: yes; 1: no)	%	0.316	0.263	0.581	0.59
	Balanced Average Score (St.dev)		0.469 (0.018)	0.418 (0.02)	0.565 (0.023)	0.564 (0.025)
Social network	Average distance to nearest market (mile=number)	miles	0.408	0.41	0.445	0.150
	Households without/not receiving social/gov help (0: yes; 1: no)	%	0.039	0.211	0.905	0.896
	Households without/not receiving community help (0: yes; 1: no)	%	0.145	0.224	0.785	0.77
	Without gov support during COVID-19 (0: yes; 1: no)	%	0.132	0.066	0.864	0.91
	Without private support during COVID-19 (0: Yes; 1: no)	%	0.618	0.25	0.612	0.68
	Balanced Average Score (St.dev)		0.268 (0.048)	0.232 (0.025)	0.722 (0.038)	0.681 (0.062)
Health	Average distance to health facilities (miles=number)	miles	0.459	0.52	0.355	0.346
	Average time to health facilities (minutes=number)	Minute	0.355	0.388	0.247	0.243
	Households without sanitary latrine/toilet (0: yes; 1: no)	%	0.026	0.105	0.085	0.185
	Household with missed work or school due to illness (1: yes; 0: no)	%	0.053	0.276	0.560	0.572
	Households with members having chronic illness (1: yes; 0: no)	%	0.053	0.118	0.303	0.365
	Households not receiving good health facilities (0: yes; 1: no)	%	0.184	0.197	0.279	0.212
	Balanced Average Score (St.dev)		0.188 (0.03)	0.267 (0.027)	0.305 (0.026)	0.321 (0.024)
Food	Household food strategy—take loans (1: yes; 0: no)	%	0.118	0.237	0.828	0.867
	Household food strategy—sell property (1: yes; 0: no)	%	0.066	0.211	0.535	0.467
	Households consume fewer foods (1: yes; 0: no)	%	0.184	0.263	0.263	0.244
	Households did not save food (0: yes; 1: no)	%	0.158	0.224	0.694	0.775
	Households did not save seed (0: yes; 1: no)	%	0.289	0.145	0.586	0.653
	Households consume non-cash food items (1: yes; 0: no)	%	0.316	0.211	0.578	0.523
	Average food insecure months (months=number)	Number	0.026	0.211	0.848	1.254
	Balanced Average Score (St.dev)		0.165 (0.015)	0.215 (0.005)	0.619 (0.034)	0.683 (0.034)
Water	Households report water conflict problems (1: yes; 0: no)	%	0.474	0.316	0.096	0.059
	Slept very few hours due to water collection duty (1: yes; 0: no)	%	0.237	0.289	0.117	0.153
	Reduced time for daily work/income generative activities due to water collection (1: yes; 0: no)	%	0.263	0.355	0.147	0.222
	Reduced time for studies/missed school due to water collection (1:yes; 0:no)	%	0.237	0.303	0.061	0.086
	Collected water from an undesirable/dirty source (1: yes; 0: no)	%	0.224	0.184	0.152	0.162
	Households that do not have private water facilities (0: yes; 1: no)	%	0.25	0.25	0.56	0.626
	Climate and hydrological changes adversely affecting households and their economy (1: yes; 0: no)	%	0.724	0.579	0.979	0.973
	Balanced Average Score (St.dev)		0.344 (0.027)	0.325 (0.018)	0.302 (0.049)	0.326 (0.049)
Natural Hazards	Household report no early warning (0: yes; 1: no)	%	0.066	0.342	0.316	0.36
	Loss of family members due to disasters and/or incurring injury (1: yes, 0:no)	%	0.013	0.066	0.037	0.063
	Household loss of housing (asset) as a result of floods/disasters (1: yes; 0: no)	%	0.289	0.184	0.324	0.371
	Households with loss of farming products as a result of flooding/disaster events (1: yes; 0: no)	%	0.316	0.263	0.629	0.724
	Past 10 years. Households with water insecurity problems (1: yes; 0: no)	%	0.803	0.763	0.654	0.716
	Within 1 year. Households with water insecurity problems (1: yes; 0: no)	%	0.355	0.276	0.263	0.284
	Flood destroys farmlands with debris (1: yes; 0: no)	%	0.592	0.395	0.739	0.822
	Balanced Average Score (St.dev)		0.348 (0.04)	0.327 (0.031)	0.423 (0.036)	0.477 (0.04)

shows the balanced average scores from the indicators making up the LVI, spread across the seven components. In the following section, we consider the climate change vulnerability of men- and women-headed households for each component, with a brief discussion of each.

4.1.1. Socio-Demographic Component

As shown in Table 1, women-headed households achieve higher scores on the socio-demographic components than men-headed households, with an average score of 0.262 (±0.017) and 0.232 (±0.03) in Myanmar and an average score of 0.422 (±0.03) and 0.374 (±0.017) in Vietnam. The contributing indicator scores—households without secondary education and family members having income-earning jobs—are higher in women-headed than in men-headed households in both countries. This suggests that during the survey period, women-headed households were less likely to obtain a good education and have income from their male family members. In Vietnam, access to basic education has been more difficult to access for women farmers than for male farmers in the past several decades. The study also demonstrates that women-headed farm households have a larger percentage of children than men-headed farm households. This is consistent with the findings that children are almost exclusively reared by women, who shoulder the large burden of providing unpaid care for sick or injured children [41,42].

Additionally, women-headed households had less access to electricity compared to men-headed households. The lack of reliable and affordable electricity beyond key urban centres has meant that many communities rely on biomass for cooking and other needs [26]. The scores for this component show that vulnerable women-headed households should be prioritised to enhance socio-demographic development. Further, addressing the norms and cultures that place additional responsibilities on women-headed households is crucial to mitigating the negative impact of climate change in the Mekong region.

4.1.2. Livelihood Component

Based on the average vulnerability score for the livelihood component, in Myanmar, there is little difference between households led by men, with a score of 0.565 (±0.025), and those led by women, with a score of 0.564 (±0.023). However, in Vietnam, men-headed households with a livelihood component score of 0.469 (±0.018) are more likely to be vulnerable than women-headed households with a score of 0.418 (±0.02). Men-headed households also report that they are currently less likely to access credit (0.316) as compared to women-headed farm households (0.263). However, in Vietnam, women-headed households receive less income from off-farm jobs (0.724) than men-headed households (0.5). Unlike the livelihood score for farm households in Vietnam, in Myanmar, men-headed households receive less income from remittance transfers from their migrant family members, while women-headed households receive less non-farm and off-farm income.

Compared to male-headed households, in both countries, women-headed farm households report that they are facing increasing difficulty repaying their loans. Access to financial services is crucial for men- and women-headed farm households to invest in farming. However, in both countries, there is a need to consider possible ways for farm households to escape the debt trap and extend loan repayments with lower interest rates. This is similar to the finding of Aryal et al. [18]: Women farmers are more vulnerable to climate risks due to their subsistence-based livelihoods and access to credit. Therefore, the livelihood component indexing scores suggest the need for access to credit, the creation of non-farm and off-farm employment opportunities, and livelihood activities in Myanmar and Vietnam.

4.1.3. Social-Network Component

In terms of social network component scores, men-headed farm households had higher indexing scores than women-headed households in both countries. The COVID-19 pandemic has had differential impacts on men and women, and this indicator is included to examine the strengths and weaknesses of the social network component. In Vietnam, men-headed households report receiving less support from private institutions (0.618) and the government (0.132) during the COVID-19 pandemic than women-headed households, with a score of (0.25) and (0.066), respectively. In the case of Myanmar, we observe similar scores. These findings show the importance of an equal and effective COVID-19 response and the need for better reporting of data by sex and social status [43]. Access to markets for selling farm products is also essential for social well-being and community income generation. In the case of Myanmar, men-headed households report an average distance to the nearest market that is slightly higher than for women-headed households. Better access to market information can be ensured by enhanced social capital and information-sharing platforms [10,43]. For instance, women with less access to resources, such as information, credit, and market services, are less likely to adopt climate change adaptation strategies, increasing vulnerability to climate risks [17,18]. Therefore, the scores for this component highlight the need for more concerted action and emergency COVID-19 responses from public and private organisations to meet the needs of vulnerable farm households in the coastal areas of Myanmar and Vietnam.

4.1.4. Health Component

In terms of the health component, women-headed households had higher scores than men-headed households in both countries. In Myanmar, the balanced average health indicator scores indicate that women-headed households, with a score of 0.321 (±0.024), are more vulnerable to climate-related natural disasters than those headed by men, with a score of 0.305 (±0.026). The same finding is observed in Vietnam, with a score of 0.267 (0.027) for women-headed households and a score of 0.188 (0.03) for men-headed households. This study reveals that family members in women-headed households suffer from more chronic health conditions (with a score of 0.118 in Vietnam and 0.365 in Myanmar) than those in men-headed households (with a score of 0.053 in Vietnam and 0.303 in Myanmar). This is because men-headed households take more initiative and are more active in utilising the health care services provided than women-headed households [44]. Women-headed households in Vietnam are also less likely to access good health facilities than men-headed households and are thus more likely to suffer from water-related diseases in both countries. WHO [43] and World Bank Group [45] highlighted that women are especially vulnerable regarding access to hygienic water and basic health facilities. This study highlights the importance of providing efficient and effective healthcare services and facilities to all people, especially children and the elderly, in the coastal areas of Myanmar and Vietnam.

4.1.5. Food Component

The vulnerability of farm households is also assessed using a seven-indicator food component. The balanced average indicator scores show that in Myanmar and Vietnam, women-headed households are more vulnerable than men-headed households. For instance, with respect to food, women-headed farm households, with a score of 0.215 (±0.005), are more vulnerable than men-headed farm households, with a score of 0.165 (±0.015). Women-headed households in both countries also report that they can no longer retain seeds for the next growing seasons and food for family members in the off-season, instead having to sell all their farm products after harvesting. This study thus shows that women-headed farm households are vulnerable to the adverse effects of climate change due to a lack of food for their families and seeds for the next growing seasons. This is in line with existing findings that farm households who do not keep food or save seeds are the most vulnerable to climate shocks in the next growing seasons [6,10].

The average food-insecure months is higher for women-headed households, with a score of 0.266 in Myanmar and a score of 0.211 in Vietnam. Men-headed households reported food insecurity conditions with a score of 0.154 in Myanmar and 0.026 in Vietnam, respectively. In food insecure situations, farm households reported that they had to rely on non-cash food items such as fishing, eel collecting, foods from forests, etc. In Vietnam, the women-headed households report that their food strategies are mostly selling their household’s properties, such as draught cattle and motorcycles, with an average score of 0.211 in women-headed households, compared to men-headed households, for which the average score is 0.066. Also, this study observes that the family members consumed fewer foods in women-headed households (0.263) compared to men-headed farm households (0.184). The food security strategies of men-headed households in Myanmar differ from those in Vietnam. Men-headed households are less likely to use loans as a means of securing food (0.828) than women-headed households (0.867). On the other hand, women-headed households exercise other options, such as selling gold and draught cattle. Therefore, the indexing scores reveal the need to address food insecurity issues and provide an affordable supply of inputs such as rice seeds, particularly to women-headed households in the region.

4.1.6. Water Component

Men-headed households reported more water conflict issues on the farm (0.474) than women-headed households (0.316) in Vietnam. The finding is similar for Myanmar, with an indexing score of 0.096 for men-headed households and 0.059 for women-headed households. It seems likely that men-headed households report more water conflict as aquaculture operations are mostly carried out by men, and it requires a higher level of technical expertise, knowledge, and capacity. Furthermore, aquaculture and agriculture livelihood activities are the major sources of water conflict issues among farm households in both countries. Aquaculture is impacted by water contamination that is caused by chemical residues and runoff produced during agricultural activities. In addition, aquaculture households sometimes have conflicts in the management of wastewater discharge within the communities, and these conflicts are frequently observed in men-headed households. Women-headed households in Myanmar and Vietnam also report sleeping less than men-headed households to allow time for water collection. We show that, compared to households headed by men, women-headed households have reduced time for daily work or income-generation activities due to fetching water for household consumption. Furthermore, climate and hydrological changes have adversely affected the income generation and social activities of the farm households surveyed. Men-headed households report that their livelihood activities are being impacted by hydrological change and climate change (0.724 in Vietnam and 0.979 in Myanmar) at a higher rate than women-headed households (0.579 in Vietnam and 0.973 in Myanmar). Therefore, the indicator scores in this water component suggest that additional action is required to address water insecurity issues and ensure affordable, clean, and hygienic water for all farm households in both countries.

4.1.7. Natural Hazard Component

The natural hazard component is considered an exposure dimension of the LVI. This component’s average scores indicate that women-headed households in Myanmar are more vulnerable than men-headed households. The average score on this dimension for men-headed farm households is 0.423 (±0.036) and 0.477 (±0.04) for women-headed households in Myanmar. However, men-headed households in Vietnam are more vulnerable, with an average score of 0.348 (±0.04), than women-headed households, with a score of 0.327 (±0.031). Factors such as a loss of assets and farm equipment during climate-related disasters are major factors contributing to the vulnerability of men-headed households in Vietnam.

Conversely, in Myanmar, all indicators contributing to these subcomponent scores indicate the higher vulnerability of women-headed than men-headed farm households. This study reveals that women-headed farm households in Myanmar are increasingly vulnerable to climate-related natural disasters as they are generally less likely to adapt or cope with the adverse impacts of climate change and natural hazards. Moreover, women-headed households also report limited (or no) access to early information and warning systems to cope with climate-related natural disasters. Aryal et al. [18] note that due to their limited access to resources and climate information, women are disproportionately impacted by climate change. Most men- and women-headed households in both countries report experiencing water insecurity over the previous ten years. Also, men-headed households report an increasing occurrence of flooding and saltwater intrusion in the region. Due to flood inundation, both farm households report that their farmlands have been destroyed by debris from floods, and they lack the capacity to cope with flooding due to their lack of access to early warning information. Therefore, there is a need for projects or programmes to improve access to early warning information and boost awareness among farm households. In addition, tackling the problems of water insecurity is also a critical strategy for dealing with the adverse effects of climate-related natural disasters in Myanmar and Vietnam.

4.2. IPCC-Defined Livelihood Vulnerability Index Assessment

The scores for the major components are presented in spider diagrams, and the contributing factor or dimensional scores are presented in a triangular diagram (see Figure 4 and also Figure A1). In Vietnam, women-headed households are more vulnerable in three out of the seven major components: health, food, and socio-demographic factors. Therefore, women-headed households are especially vulnerable in terms of their sensitivity and adaptive capacity. Conversely, men-headed households in Vietnam are more vulnerable than women-headed households in respect of the livelihood, social, water, and natural hazard components. The total calculated vulnerability score indicates that women-headed farm households are slightly more vulnerable (0.292) than men-headed farm households (0.288) in Vietnam. In Myanmar, women-headed households are more vulnerable for five out of seven components, contributing to a higher index score for vulnerability (0.476) than men-headed households scores (0.459). In Myanmar, the IPCC-defined dimensional indexing score is also higher for women-headed than for men-headed households in the exposure dimension, possibly due to their high sensitivity to the impacts of climate-related natural disasters. Therefore, this study shows that women-headed farm households are more vulnerable according to the vulnerability indexing scores than men-headed farm households in Myanmar and Vietnam.

The IPCC-defined livelihood vulnerability indexing scores of farm households in Myanmar and Vietnam are also presented in Figure 5. Overall, IPCC-defined LVI scores indicate that women-headed households (−0.472) are more impacted by climate-related natural disasters than men-headed households (−0.433) in Vietnam. Our findings are similar to those for Myanmar, where the scores for women-headed farm households (−1.414) suggest these are more affected by climate-related natural disasters than men-headed farm households (−1.394). Therefore, we suggest that it is imperative to strengthen the adaptive capacity of farm households and address the adverse effects of climate change on the sensitivity dimension of farm households, particularly regarding health, food, and water. More concerted efforts are required in both countries to improve the adaptive capacity of farm households through market innovation, investment in soft and hard infrastructures, sharing of early warning information, and affordable supply of clean water.

Table 2. Comparison of the degree of LVI of men- and women-headed households in Myanmar and Vietnam.

LVI Components	Exposure 1	Adaptive Capacity 2	Sensitivity 3	Over LVI 4
VN-Men (226)	Moderate	Moderate	Moderate	Moderate
VN-Women (74)	Moderate	Moderate	Moderate	Moderate
MM-Men (377)	High	Very High	High	High
MM-Women (222)	High	Very High	High	High

^1,2 Low (<0.3), moderate (0.3–0.4), high (0.4–0.55), and very high (>0.55); ^3,4 low (<0.2), moderate (0.2–0.35), high (0.35–0.5), and very jigh (>0.5).

explains the degree of vulnerability of men- and women-led households to climate-related natural disasters based on the IPCC-defined LVI scores. Compared to those in Myanmar, farm households in Vietnam are moderately vulnerable to climate-related natural disasters. This could be due to the mid-level scores that farm households in Vietnam achieved for the exposure and sensitivity dimensions; adaptive capacity also achieves a moderate score.

By contrast, farm households in Myanmar are highly vulnerable to climate-related disasters like flooding and saltwater intrusion. This is because livelihood activities such as fish farming and rice farming are also highly sensitive to climate change. Farm households in Myanmar have a very high adaptive capacity, but their overall LVI score indicates that they are vulnerable to the impacts of climate-related natural disasters due to the high scores on the exposure dimension. This exposure results in the sensitivity of the farming sector in both countries and the continued vulnerability of farm households. Therefore, these findings are consistent with the previous finding that farm households are adjusting their farming systems and improving their capacities if they face many external shocks or natural disasters [6].

In Myanmar, greater exposure to natural hazards leads to greater sensitivity of the farm household, even though there is little difference in the adaptive capacity dimension of households headed by men and women. Therefore, to reduce or limit the impacts of climate-related natural hazards, it is critical to pay attention to the indicators used in the exposure dimension of the study, such as the early provision of weather and climate change information, and to address the water insecurity challenges faced by farm households in the coastal areas of Myanmar and Vietnam.

5. Conclusions

In this study, we have examined the difference in climate change vulnerability of men-headed and women-headed households in Myanmar and Vietnam by using the LVI approach. This study assesses the climate change vulnerability of farm households in selected case-study areas in Myanmar and Vietnam using an LVI comprising 46 selected indicators during the period 2021–2022. Although the individual indicators for the LVI’s major components show different vulnerability trends for men- and women-headed farm households, the overall balanced indexing scores indicate that in Myanmar and Vietnam, women-headed farm households are more vulnerable to climate-related natural disasters than men-headed farm households. The low adaptive capacity of women-headed households explains their high vulnerability, even though those of exposure dimension are weak. This will lead to increased sensitivity of farm households to climate-related natural disasters in the study areas. Following this assessment, we conclude that the lower adaptive capacity of farm households leads to greater sensitivity to climate change when external shocks like flooding and saltwater intrusion affect agriculture and aquaculture farming households in the coastline areas of Myanmar and Vietnam.

As vulnerability is time- and location-specific, the vulnerability of farm households in the study areas can fluctuate over time or within a different period, and thus, additional assessments can be conducted over a different time and or longer period. This study provides a snapshot of the context and location-specific information for policymakers and development planners in these areas. We, therefore, recommend additional assessments in different regions of Myanmar and Vietnam using additional indicators such as impacts on the ecosystem, biodiversity, etc., as well as advanced vulnerability assessment methods to gain a deeper understanding of the relative climate change vulnerability of men- and women-headed households in Myanmar and Vietnam. In addition, the adaptive capacity of farm households in both countries, particularly women-headed households, should be strengthened through the creation of livelihood income opportunities, provision of formal and informal training programmes, and early warning information, as revealed in this study. Reducing the sensitivity of farm households requires additional investment in both soft and hard infrastructures, market innovation and information sharing, equitable supply of farm inputs and foods, as well as improved management of water for clean and affordable supply in the coastline areas of Myanmar and Vietnam. Moreover, there is a need for private and public organisations to make a more concerted effort to improve the adaptive capacity of farm households.