Climatic changes and vulnerability of household food accessibility : A study on Malaysian East Coast Economic Region

Food security is a critical global aspect of the individual and household well-being regardless of whether it is viewed locally, globally or within communities. Hunger and food insecurity are detrimental to the basic rights of a human being and are a prelude to problems including health, nutrition and development. The household food security is a more current understanding of the food security concept. Most of the literature in this subject from the 1980s assumes national food security to be similar to food self-sufficiency and this poses a few challenges. According to Debebe (1995), household food security represents factors that are linked to the actions of “[…] acquisition, household procurement strategies, and socio-economic condition of the society”.

Various types of indicators are required to measure the many different aspects of food security at an individual, household and country levels (Hoddinott, 1999). Various correlated environmental, socioeconomic and political aspects that need multi-pronged approaches have an influence on food security (Debebe, 1995). Negatu (2006) suggests that one of the factors that has an effect on food security is land tenure insecurity and the other is not having accessibility to productive resources, low educational level, reduction in land holdings, not having accessibility to proper transport structure, low livestock productivity, unaware of suitable interventions, bad storage technology and a high unemployment level. Other factors that also have an effect on food security include crop pests, drought and sudden torrential rainfall; not having a draught oxen; increased population; and shortage of land (Negatu, 2006).

Natural indicators result in specific higher variation in production costs and production that are linked to irrigation, infection and insecticide treatment and higher losses (Lovendal and Knowles, 2006). Depending on the capability of traders bringing in food from various other areas and the available stock levels, this could potentially disrupt food supply and cause prices to increase further. The main environmental indicators prove to be trends and not shocks that occur. They negatively affect production through reduced soil fertility and lower yields and through the rise of production cost per unit. The profitability of farming and the potential to gain income from other resource-based activities are limited.

There are four dimensions of food security such as availability of food, stability of its supply, accessibility to food and utilization of food (FAO, 2005; 2008a). A vulnerable food system occurs when any dimension of food security is insecure and uncertain. Understanding the vulnerability of the factors of food security can assist in the identification and comprehension of the fundamental dimension of a population’s well-being and to recognize the most vulnerable communities in the regions or subgroups.

As other dimensions, household food accessibility is also very important for household food security, which indicates household access to sufficient and suitable food for a nutritious meal that households can gain access based on the legal, economic, political and social arrangement of a community in which they live (including conventional rights such as access to normal resources). Garrett and Ruel (1999) point out that a household’s accessibility to food is dependent on the income of the household to buy food at the current prices or if they have enough land and resources to produce the necessary food, besides receiving aid through prescribed programs or other unofficial networks during times of need. Other factors besides the issues of price and income can also have an effect on the household’s caloric availability. These include location, educational level of household members and the demographic structure of the household (Garrett and Ruel, 1999). Nyariki and Wiggins (1997) suggest that food access in a household is implicitly dependent on the capability to use physical, natural and human resources efficiently. It includes the availability of resources such as labor and land, which has a significant role in the production of food and the ability to use these resources sufficiently, is the determinant of food security in the household.

According to Lovendal and Knowles (2006), food utilization and food accessibility are mainly affected at the household level by economic, social, health and life cycle-related indicators. Being dependent on subsistence farming with no other income-generating activities or no accessibility to markets can lower the availability of food among households. Life events such as weddings and funerals often incur a large amount of additional expense, thus also lowering the capability for purchase of food. Likewise, disability, injuries and illnesses incur extra expenses and lower the supply of labor and income, thus causing insufficient utilization of food. Economic indicators are able to lower food accessibility via the loss of income by being unemployed or because of the failure of income-generating activities such as farming, small manufacturing or trade.

Malaysia is a fast-growing developing country. Therefore, there are scopes of frequent changes in the macro variables that would affect food security. Moreover, climatic factors are changing rapidly in Malaysia and have been having adverse impacts on food production (Alam et al., 2010, 2011). As such, household food insecurity is not only related to social and economic factors, but is also statistically and significantly linked to the direct and indirect impacts of climatic factors (Alam et al., 2016a). Previous studies show that 50 per cent or more of the low-income rural households face some level of food insecurity with 34.5 per cent child hunger (Shariff and Khor, 2008). The percentage of households with food insecurity in Malaysia is higher in urban low-income households, such as 66.6 per cent (Zalilah, 1998) and 65.7 per cent (Zalilah and Ang, 2001), compared to rural low-income households at 58 per cent (Shariff and Khor, 2008). Another recent study shows that among the poor and low-income households, 23.3 per cent are mildly food insecure, 14.3 per cent are moderately food insecure and 9.6 per cent are severely food insecure in Malaysia (Alam et al., 2016b).

Therefore, to understand the level of vulnerability of household food security, this study is an attempt to check food accessibility – one of the important dimensions of household food security, and its linkage to climatic changes in Malaysia.

Climate changes affect the fundamental aspects of food production negatively, which include water, soil and biodiversity. In a broader viewpoint, it has an effect on all the four types of food security’s dimensions (FAO, 2008a). FAO also highlights the impacts of the components of climatic changes on household food accessibility that are given in Table I.

In general, climate changes are expected to cause the number of people at risk from hunger to grow (FAO, 2005). In the case of rural people who tend to produce a large portion of food for own consumption, the effects of climate changes on food production may lower the availability to the extent that choices of allocation would have to be determined inside a household. Similar choices will be faced by the rural and urban non-farming low-income households whose incomes are below the poverty line due to the effects of climate changes. Urbanization is a worldwide phenomenon and a large number of the growing population in the urban areas are considered poor (Ruel et al., 1998). Therefore, over time, the allocation problems due to climate changes would potentially become more prevalent in the urban areas compared to those of the rural.

Most food is not produced by individual households but received via borrowing, buying and trading (Du Toit and Ziervogel, 2004). Effects of climate changes on income-earning activities can have an effect on buying food, and climate changes or extreme climates may impact the availability of certain foods, which may affect the price. The high prices may cause some food to be unaffordable and can affect the nutrition and health of an individual.

The price of food may have an effect on households’ food accessibility, by putting a constraint on the acquisition of suitable food for a balanced diet and the buying power of food aid schemes. Climate variations will probably cause a substantial rise in food prices (von Braun, 2007; Cline, 2007). Temperature rise of higher than 3°C may result in prices to go up by 40 per cent (Easterling et al., 2007). For most countries, the ratio of the cost of daily minimum food basket to the daily income average is utilized as a method to calculate poverty (World Bank PovertyNet, 2008). If this ratio goes down lower than the fixed threshold, it shows that food is affordable and people can afford it; if it goes higher than the fixed threshold, it means that the food is not affordable and people will face trouble trying to get sufficient food. This criterion indicates critical poverty and can be utilized to decide if people are experiencing temporary food insecurity, due to decreased food supply and higher prices and/or due to a sudden dip in the household’s income.

Plant and animal diseases and pests, as well as invasive alien aquatic species, related to climate changes will lower the quantity of food availability, affect the production system’s stability and lower food accessibility by reducing the income gained from livestock, lower cash crops and food yields, reduce forest productivity and the aquatic populations, thus increasing the control costs (FAO, 2008b). The price of food will have an effect on food accessibility in households, by limiting the purchase of suitable food for a nutritious and balanced diet and to gain from food aid programs.

The distance that food has to travel to get to the table of the consumer has an impact on its price (Pretty et al., 2006). Rise in fuel price could cause a higher cost of food, which in turn, can increase food insecurity. Biofuels are a growing market, which is anticipated to affect food security, because crops produced for feedstock as liquid biofuels can take the place of food crops, which have to be sourced from another place at a higher price.

Changes in demand for agricultural labor that is seasonal, due to changes in production caused by climate changes, can have an effect on income-generating ability either negatively or positively. The requirement for seasonal labor may decrease in many places due to mechanization, and demands for labor are often lowered with crop failure, mostly due to flood, drought, pest outbreaks or frost caused by climate changes. However, several adaptation choices may improve the demand for seasonal labor in agriculture. In most regions in the world, local food prices are strongly affected by conditions of the global market, but there may be short-term variations associated with changes in national yields, which are affected by climate changes, besides other factors. Food price increase has a real effect on income, with low-income households often facing the most suffering, as they normally devote larger portions of their incomes to food compared to higher-income households (Thomsen and Metz, 1998).

Food preferences ensure the types of food a household buys. Climate changes may have an effect on the economic and physical availability of some of the food preferred and this might be the possibility of purchasing the preferred food. The changes in the price and availability of the main food variety may cause the households to change their normal food basket or use more of their total income to purchase some of the food when the prices have gone up. Food affordability is dependent on the association between the cost of a normal food basket and the household’s income. Global food markets may show higher price volatility, disrupting the stability of profits for farmers and having access to purchased food for farming and non-farming poor households. Food emergencies, which increase the instability of supplies, due to climate changes, will probably increase the magnitude and frequency of food emergencies for which the global food system is not equipped to manage. Increased human conflicts, created partly due to migration and competition for limited resources also due to changing climatic conditions, could cause destabilization of food systems at all levels. Changes in climate will expand conflicts in many ways; however, these relationships must be addressed carefully. The higher frequencies of drought may cause communities to migrate to a new place, which can create conflict among people fighting for the same resources in the new place. Limited resources could also instigate conflicts and be the result of environmental changes globally.

Data for this study were collected through an extensive questionnaire survey at the household level in the East Coast Economic Region (ECER) in Malaysia. ECER is more than half of Peninsular Malaysia with the area of about 66,000 km² that covers the states of Kelantan, Terengganu, Pahang and Mersing in Johor. ECER is crucial for two major reasons:

1. ECER is the most vulnerable area in Malaysia because of climatic changes.

2. The income level of this region is low and poverty rate is high, which is an obstacle to achieve Vision 2020 (Alam et al., 2012).

This study is based on primary data collected from 460 households in the urban and rural areas of ECER. The urban area of Kuantan and rural area of Pekan were selected from the Pahang state. The urban area of Kota Bharu and the rural area of Tumpat were chosen from the Kelantan state. The urban area of Kuala Terengganu and the rural area of Maring were taken from the Terengganu state. A questionnaire guide technique was used to collect the data. The survey was conducted by the regular enumerators of the Implementation Coordination Unit (ICU) from Pahang, Kelantan, and Terengganu states in the months of July-October 2012.

The survey follows the two-stage cluster random sampling technique. Initially the samples are clustered by location and then by poverty category. Finally, from each category, samples are picked randomly from the E-Kasih database. The E-Kasih system is an integrated database system that has enlisted poor households at the national level to plan, implement and monitor poverty programs in Malaysia.

There are many factors relevant to the vulnerability of household food accessibility due to climatic changes and non-climatic/general changes. To measure the degree of vulnerability of these factors and their linkage with climatic changes, the perceptions of the households have been shown in three forms:

1. measures the current level and five year’s earlier level of these factors by using the five-point Likert scale of 1 (very low) to 5 (very high);

2. measures the changes in the factors by comparing the current level with a level of five years earlier; and

3. measures the level of climatic influences on the changes of these factors over the five-year period.

This study also examines the equality of two means between the current status of the relevant factors and five years’ earlier factors through the ANOVA F statistics for chi-square (χ²) two-sample test:

Here:

O_i = current status of i-th factors (ordinal data).

E_i = five years’ earlier status of i-th factors (ordinal data).

The variables that are considered in this study are as follows:

• X1 Prices of general food items.

• X2 Difference between rural and city food prices.

• X3 Price differences among shops in the local market.

• X4 Household food shortage caused by high food prices.

• X5 Stability of food price in market.

• X6 Level of household income.

• X7 Household food shortage caused by low level of income.

• X8 Household budget arrangements to buy food anytime.

• X9 Household access of quick credit to buy food.

• X10 Availability of discount or offer on food price in the local market.

• X11 Efficiency of household food acquisition or collection system.

• X12 Household dependencies on common resources for cattle or livestock feeding.

• X13 Competition among people for common resources.

• X14 Food and medicine expenditure for cattle and livestock.

• X15 Household food security-related supports from agencies.

Among the respondents of the study, around 60 per cent are selected from the urban area and 40 per cent from the rural area. In total, 69.8 per cent are male and 30.2 per cent are female respondents. The male respondents are 69.7 per cent among the urban respondents and 69.8 per cent among the rural respondents. Males are the majority group, and the respondents represent the household data.

The average age of all respondents is 50.7 years, where that of male is 48.9 years and that of female is 54.7 years. Among the male respondents, the maximum age is 83 years and the minimum is 26 years. Among the female respondents, the maximum age is 83 years and the minimum age is 25 years. In total, only 3.7 per cent are aged below 30 years and 23 per cent are aged above 60 years. The maximum number of respondents is in the age range of 30-40 years at 35.2 per cent of the total. Exception is only for the rural area of Kelantan, where the maximum number of participants is in the age group of 50-60 years.

Most of the respondents have secondary education (46.5 per cent) followed by basic primary education (35.9 per cent). Only 3 per cent of the respondents have certificate/diploma-level education and 14.6 per cent are illiterate. The numbers of illiterate people are higher in the urban area at 17.3 per cent, whereas in the rural area, it is at 12.7 per cent.

Among the total respondents, 72.8 per cent are married, whereas 3 per cent are unmarried. In total, 24.2 per cent of the respondents are found to be either widowed or divorced. Most families (20.7 per cent) have five persons in the household and average family members is 5.6 persons per household. There are one to three persons available in 17.9 per cent of the households. There are eight or more people available in 17.9 per cent of the households.

Among the respondents, most of the households (63.3 per cent) have no children (aged below 6 years) and the average number of children (aged below 6 years) per household is 0.6. Majority of them have only one child, and their ratio is 20.9 per cent. Most of the households (24.1 per cent) have no school-going children (aged below 6 years) and the average number of school-going children per household is 2.2. Families that have two to three school-going children are 40.3 per cent.

Among the households, 83.7 per cent have their own residence. This proportion is same for both urban and rural households. In the rural area, 50.9 per cent of houses are made up of mixed types (using both brick and wood) of houses. In the urban area, 34.6 per cent of people live both in mixed and wood-made houses.

The respondents are engaged in various types of jobs, such as babysitter, batik designer, building contractor, small business, bus driver, canteen assistant, car service, carpenter, chef, clerk, cleaner, construction worker, craftsman, electrical technician, van driver, fisherman, fruit dealer, furniture shopkeeper, gardener, jewelry shopkeeper, hawker, daily labor, farmer, mason, painter, plumber, saloon worker, security guard, imam, teacher, government servant, tailor, taxi driver, welding worker, factory laborer, etc. In 90 per cent of the cases, the heads of the households do not have any supplementary job. The rural households have more supplementary job options than urban households. The spouses of most of the households (88.3 per cent) are not engaged in any job. However, the number of spouses of urban households engaged in jobs is more compared to that of the rural households.

Around 76.3 per cent of the households have only one earning member; 15.7 per cent have two earning members, and 5.4 per cent have three earning members. The proportion is almost the same for the urban and rural areas. At the household level, on an average, the earning family members to total family members’ ratio is 0.287, indicating that, on average, only 28.7 per cent of the family members are engaged in economic activities. Among the surveyed households, in 47.2 per cent of the cases, this ratio is up to 0.2, and in 3.7 per cent of the cases, the ratio is 1 which means a single-person family.

The household income of most of the respondents is between RM 5,001-1,000 range. Further, 35 per cent of the household’s monthly income is up to RM 500, which is more in the urban area in comparison to that of the rural area. Only 3.6 per cent of a household’s income is above RM 2,000, which is more in the rural area in comparison to the urban area. The per-capita income of most of the respondents is between RM 100-200. The per-capita income of 59.3 per cent of respondents is up to RM 200.

According to the survey findings, around 55 per cent are still poor and hard core poor. Around 9 per cent of the people have an income of more than RM 400. The people in the low household income group also belong to the low per-capita income group.[¹]

The distribution of income among the surveyed household is not that wide. However, here, the top 10 per cent earn 23.2 per cent of the total income, while the bottom 10 per cent earn only 3.4 per cent of the income, and the top 20 per cent earn 36.9 per cent of the total income while the bottom 20 per cent earn only 8.8 per cent.

The degree of inequality is measured by the inequality ratio, also known as the Kuznets’ ratio. The ratio measures inequality through the ratio of income of top 20 per cent to that of the bottom 40 per cent. Here, the ratio is 1.624 [(1 − 0.631)/0.227], meaning that the degree of inequality at household level is more than 1.5 times. Moreover, if the inequality ratio is measured in terms of population instead of household, the degree of inequality is roughly 1.405 (0.358/0.255). The degree of inequality at the population level is less than the household level because there are likely less population in the poorer households and more population in the richer households.

The most vulnerable groups are those who have no household saving, and as a result, their per-capita savings are nil. Among the respondents, around 82 per cent do not have any savings. Therefore, any type of spike in regular life due to natural disasters, disease, crime or social issues will push them toward borrowing. The rural households save less compared to the urban households.

The changes in the prices of general food items (X1) in Malaysia over the past five years are found statistically significant at 1 per cent level (Table II). Around 81.5 per cent of the respondents mention the changes are positive and overall, the average score increases from 2.5 to 3.9 (Table III). It indicates that the vulnerability of the prices of general food items in Malaysia has increased over the time.

The changes in the difference between the rural and city food prices (X2) in Malaysia over the past five years are found statistically significant at 1 per cent level (Table II). Around 58 per cent of the respondents mention that changes are positive and overall the average score increases from 2.75 to 3.49 (Table III). It indicates that the vulnerability of the prices of general food items in Malaysia has increased over time.

The changes in the price differences among shops in the local market (X3) over the past five years are found to be statistically significant at 1 per cent level (Table II). However, only 40.9 per cent of the respondents mention the changes are positive, and overall, the average score increases from 2.85 to 3.29 (Table III). It indicates that the vulnerability of the price differences among shops in the local market has increased over time.

The changes in the household food shortage caused by high food prices (X4) in Malaysia over the past five years are found to be statistically significant at 1 per cent level (Table II). However, only 46.7 per cent of the respondents mention that the changes are positive, and overall, the average score increases from 2.5 to 3.03 (Table III). It indicates that the vulnerability of the household food shortage caused by high food prices in Malaysia has increased over time.

The changes in the stability of food price in the market (X5) over the past five years are found to be statistically significant at the 1 per cent level (Table II). Around 78 per cent of the respondents mention that the changes are positive, and overall, the average score increases from 2.58 to 3.69 (Table III). It indicates that the vulnerability of the stability of food price in market has decreased over time.

The changes in the level of household income (X6) in Malaysia over the past five years are found to be statistically significant at the 1 per cent level (Table II). However, only 33.5 per cent of the respondents mention that the changes are positive, and overall, the average score increases from 2.63 to 2.34 (Table III). It indicates that the vulnerability of the level of household income in Malaysia has increased over time.

The changes in the household food shortage caused by low level of income (X7) in Malaysia over the past five years are found to be statistically significant at the 1 per cent level (Table III). However, only 33.5 per cent of the respondents mention that the changes are positive, and overall, the average score increases from 2.56 to 2.31 (Table III). It indicates that the vulnerability of the household food shortage caused by a low level of income in Malaysia has increased over time.

The changes in the household budget arrangement, ready-to-buy food anytime (X8) in Malaysia over the past five years are found to be statistically significant at the 1 per cent level (Table II). However, only 31.3 per cent of the respondents mention that the changes are positive and overall, the average score increases from 2.67 to 2.37 (Table III). It indicates that the vulnerability of the household budget arrangement, ready-to-buy food anytime in Malaysia has decreased over time.

The changes in the household access of quick credit to buy food (X9) over the past five years are found to be statistically significant at the 1 per cent level (Table II). However, only 27 per cent of the respondents mention that the changes are positive, and overall, the average score increases from 2.13 to 2.32 (Table III). It indicates that the vulnerability of the household access to quick credit to buy food has decreased over time.

The changes in the availability of discount offer on food price in the local market (X10) in Malaysia over the past five years are found to be statistically significant at the 1 per cent level (Table II). However, only 42 per cent of the respondents mention that the changes are positive, and overall, the average score increases from 3.05 to 2.45 (Table III). It indicates that the vulnerability of the availability of discount offer on food price in the local market in Malaysia has decreased over time.

The changes in the efficiency of household food acquisition or collection system (X11) in Malaysia over the past five years are found to be statistically significant at the 1 per cent level (Table II). However, only 33.3 per cent of the respondents mention that the changes are positive, and overall, the average score increases from 2.66 to 2.32 (Table III). It indicates that the vulnerability of the efficiency of household food acquisition or collection system has decreased over time.

The changes in the household dependency on common resources for cattle or livestock-feeding (X12) in Malaysia over the past five years are found to be statistically significant at the 1 per cent level (Table II). However, only 25.2 per cent of the respondents mention that the changes are positive, and overall, the average score increases from 2.15 to 2.03 (Table III). It indicates that the vulnerability of the household dependency on common resources for cattle or livestock-feeding in Malaysia has increased over time.

The changes in the competition among people for common resources (X13) in Malaysia over the past five years are found to be statistically significant at the 1 per cent level (Table II). However, only 31.7 per cent of the respondents mention that the changes are positive, and overall, the average score increases from 2.49 to 2.14 (Table III). It indicates that the vulnerability of the competition among people for common resources in Malaysia has increased over time.

The changes in the food and medicine expenditures for cattle and livestock (X14) in Malaysia over the past five years are found to be statistically significant at the 1 per cent level (Table II). However, only 25.2 per cent of the respondents mention that the changes are positive and overall, the average score increases from 2.13 to 1.87 (Table III). It indicates that the vulnerability of the food and medicine expenditures for cattle and livestock in Malaysia has increased over time.

The changes in the household food security-related supports from agencies (X15) over the past five years are found to be statistically significant at the 1 per cent level (Table II). However, only 35.9 per cent of the respondents mention that the changes are positive, and overall, the average score increases from 2.31 to 2.9 (Table III). It indicates that the vulnerability of the price household food security-related supports from agencies has decreased over time.

For all cases, most of the respondents mention that the influences of climatic factors are low for the changes of these factors of household food accessibility over the five years (Table IV).

The study finds that the vulnerability of the factors of household food accessibility has changed statistically and significantly over the past five years. The mean values of the current status of these factors of household food security differ significantly at the 1 per cent significance level from the scenario of five years ago. The vulnerability of the household food accessibility increases due to the increases in the prices of general food items, food prices inequity between rural and urban locality, price differences among shops in the local market, high prices leading to food shortage, changes in the household income, income-led food shortage, dependency on common resources for cattle or livestock feeding, competition among people for common resources and food and medicine expenditures for cattle and livestock. However, vulnerability of the household food accessibility decreases due to the increases in the stability of food price in market, household budget arrangement to buy food anytime, household access of quick credit to buy food, discount offer on food price in the local market, efficiency of the household food acquisition or collection system and the household food security-related supports from agencies. However, in most cases, climatic issues have been found to be less responsible in the changes of these factors.

There should be some initiatives taken by the government in response to these factors that increase the vulnerability of the household food accessibility. There should be a proper monitoring body to ensure a consistent supply of food at regular prices both in the rural and urban markets, especially during times of natural calamity or off-season fluctuations. At the same time, income-augmenting activities also need to be emphasized. The dependency on common resources is high for poor and low-income groups. Therefore, initiatives for common resource protection and awareness creations are also important for the sustainable management of the common resources.

Although Malaysia is adversely affected by climatic changes, the factors of the household food accessibility have not yet been remarkably affected. However, climate changes are considered as one of the major potential threats for household food security in Malaysia. Therefore, to ensure a sustainable household food security in Malaysia, the design of food security programs must integrate climatic changes into the picture, and food security approaches must recognize climate changes as an important driver. This integration would also increase household capacity to adapt to climatic changes. At the same time, it must be ensured that the climate changes adaptation approaches and strategies that reduce vulnerability to climate changes also increase the household food security.

The authors are thankful to the Ministry of Science, Technology and Environment of the Government of Malaysia for generously funding the research, under the Research University Grant Project UKM-GUP-PI-08-34-081, AP-2014-017 and FRGS/1/2012/SS07/UKM/01/3.