EXECUTIVE SUMMARY
On 25 April 2015, a 7.8 magnitude earthquake struck Nepal with its epicentre in Gorkha district, approximately 81km northwest of the capital, Kathmandu. Intense tremors, and subsequent aftershocks, landslides, and avalanches caused widespread damage to homes, land, public and private infrastructure and livelihoods, affecting millions of people across 39 out of 75 districts. The Nepalese government categorized 14 of these districts as severely affected: Dhading, Gorkha, Rasuwa, Kabhrepalanchok, Nuwakot, Dolakha, Sindhupalchok, Kathmandu, Ramechhap, Bhaktapur, Lalitpur, Makawanpur, Sindhuli and Okhaldhunga. Combined, over five million people reside in these districts.
Amid ongoing emergency relief efforts following the initial earthquake, Nepal was struck by a second earthquake on 12 May 2015, which registered a magnitude of 7.4 on the Richter scale. The epicentre of the second earthquake was located further east than the first, on the confluence of Sindhupalchok and Dolakha districts, compounding the initial devastation in already affected areas. According to government estimates, the earthquakes combined caused over 8,790 casualties and 22,300 injuries, and left over 500,000 houses and hundreds of historical and cultural monuments destroyed. It is estimated that the earthquakes affected the lives of approximately eight million people, constituting more than a quarter of the population of Nepal.
In light of this, in May, the World Food Programme, in partnership with the Government’s Nepal Food Security Monitoring System (NeKSAP) and the Food Security Cluster, conducted a food security assessment in 11 of the aforementioned districts, excluding the three districts of Kathmandu Valley. This baseline facilitated and informed the immediate emergency response.
The following joint assessment, conducted over the course of September and October 2015, was designed both as a monitoring exercise and to further inform broader early recovery efforts identified as priorities in the Post-Disaster Needs Assessment, including food security, livelihoods, agriculture, service access and protection. To that end, in addition to the partners in the May assessment, it also included more partners, including OCHA, FAO, UNDP, REACH, the Early Recovery and Protection Clusters, and the Nepal Red Cross Society.
The assessment sought to: a) update information on key thematic areas included in the Post Disaster Needs Assessment, b) identify and measure changes since the May 2015 baseline assessment, c) identify key trends, gaps and risks across the assessed sectors and d) enable all stakeholders to derive the information they need to define their own cluster or agency-specific early recovery and development strategy.
Key findings from the assessment are presented below, which have been agreed upon jointly by all partners. The results presented here are for households in 11 districts worst affected by the earthquake and capture the situation at one point in time, mid-September to mid-October, which follows the end of the lean season in August/September and the end of the monsoon in September and precedes the festival season in October/November and the harvest of summer crops in November/December. The results should be interpreted with these contextual factors in mind. Furthermore, because of the geographic focus and timing of the joint assessment the results are not generalizable to other parts of Nepal nor to other times of the year. For nationally representative data, the readers should refer to the latest Nepal Living Standards Survey and Nepal Demographic and Health Survey.
Demographics
With an average size of five individuals, Nepali households are relatively large and exhibit high rates of dependency, mostly due to the presence of children which account for a third of the population across the assessed areas. Average dependency ratios approach one, suggesting one dependent per economically active individual or caregiver. Over a fifth of households are headed by women, with over a quarter of urban households headed by women relative to under 20% of rural households. This is likely due to high levels of male out-migration, as well as potentially higher male death rates, rising divorce rates and female emancipation. When examined by self-reported caste or ethnicity, Janajati households were the most common, accounting for over half (57.7%) of households; Brahmin/Chhetri households comprised a third (32.9%) and Dalits a tenth (9.2%) of all households. It is worth noting that the results of this survey vary from other studies conducted on social composition primarily because various Janajati sub-groups were brought under the Janajati umbrella group.
More than a third (37.3%) of households were reportedly hosting vulnerable individuals, including people with disabilities, chronically ill persons and pregnant and/ or lactating women. Among these households, chronic illness was the most commonly reported vulnerability, present in nearly a fifth (18.9%) of these households, followed by pregnant and/or lactating women (12.3%) and people with disabilities (11.3%).
Migration
Reported migration rates since the earthquake were highest in Ramechhap, where 8% of individuals aged 17 or older at the time of the earthquakes had reportedly migrated away from their previous location; Sindhupalchok, where 6.5% had done so; and Gorkha, 6%.
The vast majority (82%) of all individuals who had reportedly migrated elsewhere since the earthquake were male. Approximately 4% of individuals aged 17 or over reported intending to migrate either within Nepal or abroad within the three months following the survey, indicating a continued rate of migration in the medium term at least. Of these, an estimated 84.2% are also male and aged 17 or over.
Food Security
The survey shows a significant improvement across all food security indicators since May 2015, with less than one fifth (17.7%) of all households falling below the acceptable threshold for food consumption as of September, relative to 45.9% in May.1 This improvement is likely due to several factors, including the large amount of humanitarian assistance provided, the start of the harvest of summer crops, and the restoration of access to markets and improvements to their general functionality.
Pockets of food insecurity and vulnerability do persist however, most notably in Dolakha, Sindhupalchok, Gorkha, Sindhuli and Ramechhap, where more than a fifth of all households had inadequate (poor and borderline) food consumption at the time of the assessment. Elevation was found to be a strong determinant of food consumption, with the proportion of households with inadequate food consumption at 14.6% in the low hills and 26.5% in the high hills.
Higher levels of inadequate food consumption were noted in rural areas, where a fifth (19.8%) of households were deemed food insecure, relative to only 6.4% in urban areas. Following on from this, households relying primarily on agriculture, agricultural on-farm labour and lowskilled daily labour were found to have lower food consumption scores, lagging behind other livelihood groups. Inadequate food consumption was highest amongst Dalit households, over a third of whom (33.6%) fell below the acceptable threshold for food consumption, making caste an important factor of food security status at household level.
Dietary diversity was also found to have recovered since May 2015, in line with food consumption levels. The average frequency of consumption, in days per week, of animal protein has increased by an entire day from 0.7 in May to 1.7 in September. The same is true for the consumption of vegetables, which registered a significant increase in consumption from 3.3 days per week in May to 5.8 days per week in September. Dietary diversity scores (DDS) largely reflect these consumption patterns, with two thirds (66%) of households found to exhibit high dietary diversity, a third (32.0%) a medium dietary diversity and an estimated 2% of households having a low dietary diversity.
Overall, 20.6% of households reported that they had adopted any form of food-based coping strategy, a substantial decrease from May when 68.7% of households reported doing so, suggesting that the need to use them has since diminished in line with the wider recovery in food consumption. Furthermore, fewer households were frequently adopting food-based coping strategies, with 6.7% in September compared to 15.9% in May.
Market availability of food items has generally increased since May: more than 90% of surveyed households reported increased availability of cereals, pulses, vegetables and oil in their nearest food market. A positive correlation between a household’s food market access (in travel time) and their food consumption status is observed. The nearer the food market is, the better off the households’ food consumption and dietary diversity are.
Nevertheless, a majority (84.8%) of households reported the need for food or cash assistance in the next six months, reflecting the need to continue supporting household recovery, boost purchasing power and smooth overall consumption during the upcoming winter period. Among these households, overall, the top-five reported needs (as a percentage of households) included rice (73.8%), one-off cash grant (60.5%), pulses and lentils (57%), vegetable oil (44.7%) and cash for work (25.6%).
Livelihoods
Being a labour intensive, predominantly agrarian economy, labour market participation was generally high for working age individuals aged 17 and older. Findings indicate that more than half (54.3%) of individuals belonging to this age group were reportedly engaged in some form of income-generating activity. Reported employment rates were highest in Ramechhap,
Sindhupalchok and Gorkha where an estimated 60% of individuals aged 17 and older were engaged in some form of productive labour. Participation rates were lowest in Makawanpur and Kabhrepalanchok districts, where less than half (43.4%) of surveyed individuals were working.
Agriculture was the most commonly reported current income source for both men and women, with a slightly higher proportion of households reporting this as the primary source of income for women (65%) as opposed to men (63%). The proportion of households reporting women as economically inactive (21.8%) at the time of the assessment was also much higher than the corresponding figure for men (8.3%); further, a higher proportion of women in urban areas (27.6%) were deemed economically inactive when compared to rural areas (20.7%), owing largely to the widespread practice of agricultural activities. Finally, assessed households reported that men (14%) were more likely to receive remittances than women (3%), whilst women were in turn more likely than their male counterparts to receive welfare payments. This is not to say that women did not access remittances, simply that men were the primary recipients.
Households reportedly earned an average of 12,322 Nepali Rupees (NPR) in the 30 days prior to the survey. Average reported household incomes were lowest in Dolakha (8,903 NPR), Sindhupalchok (9,204 NPR) and Okhaldhunga (9,293 NPR). Households residing in rural areas generally earn less than counterparts residing in larger urban areas. The same relationship holds for elevation, whereby incomes steadily diminish in line with increasing altitude.
Overall, an estimated 78.9% of households reported holding debt at the time of the assessment, with high outstanding debt loads across the board. On average, debt loads were reported to exceed average monthly incomes by a ratio of 24:1, indicating a high propensity for debt accumulation. In addition, rural household debt exceeded urban household debt by a ratio of 4:1, suggesting much higher rates of borrowing in rural areas. Debt levels also steadily diminished with increasing elevation, indicating constrained access to credit in high hill areas. A quarter of all outstanding debt was accumulated in the six months following the earthquake, suggesting that the ability to take on debt has been a cornerstone of the household recovery effort for many.
The majority of lost or damaged assets, as a result of the earthquake, were reportedly tools and infrastructure associated with agricultural livelihoods, which is reflected in lower expectations of agricultural production and higher debt levels. The infrastructure and assets which were reported to have incurred the most damage include livestock sheds (reported by 30.8%), produce storage facilities (21.7%), sickles (17.8%), spades (17.5%), doko baskets (16.7%) and other agricultural tools (12.8%). With the most significant and the most difficult to recover infrastructure, such as buildings and storage facilities, having incurred the most damage, this not only diminishes productive capacity, but also household wealth in the process.
Agriculture
In the high hills, 73.8% of the population surveyed relied at least partly on agriculture. Farms in these areas are typically smallholdings or less than one hectare, with the main cereal crops maize and paddy rice. In addition, potato production represents a third of staple and pulse production. For almost three quarters of these households, the lack of irrigation systems means they can only grow one crop cycle annually. Livestock raising, mostly sheep and goats, is therefore an important complement for 88% of them. At the time of the survey, 43% of these households relied on their own production for cereal consumption, 64% for milk and 82% for vegetables.
In the mid hills, 80% of the population consider themselves farmers, and 91% own livestock.
This is the most heavily agrarian region of the three. 50% of households relied on their own production for cereal consumption. While areas in the lower hills were found to have a lower share of households reporting agriculture as their main livelihood (65%) since this is a more urban region, 84% of these households reported to own livestock.
Brahmin/Chhetri households are three times as likely to sell their agricultural products (30%) than Dalits (10%), but this difference narrows down between male (22%) and female (19%) headed households. The unequal distribution of land and varying yields by ecological belt mean the vast majority of farming households do not produce enough for their household needs, and rely on additional income sources.
Farming tools were largely destroyed by the collapse of houses and landslides that followed the earthquakes and have not yet been recovered. While this most likely has a direct influence on farmers’ productivity, the physical destruction of (any) physical asset also acts as a proxy for earthquake and landslide intensity. Five months after the event,
36.6% of households surveyed reported still having damaged or unusable tools, while the most affected districts were Sindhulpalchok (79%), Rasuwa (69%) and Dolakha (67%).
Across the 11 districts, the near totality of households reported damage to their storage capacity which has not yet been recovered, and 44% reported that their facilities were almost entirely destroyed. Damage appears particularly acute in Sindhulpachok (80%), Dolakha (62%), Rasuwa (59%), and Gorkha (55%). Grain and seed storage bags are therefore among the top three priority needs for 28% of farmers surveyed. In addition, these households are particularly food insecure, as they lost their food and seeds stocks, having lower food consumption scores than others assessed.
In the 11 districts surveyed, around one third of agricultural households report having an irrigation system. This proportion varies with elevation, with households in higher hills least likely to have an irrigation system (24%) and those in low hills most likely to have one (43%).
Caste is also correlated with access to irrigation systems, in favour of Brahmin/Chhetri households (46%) over Janajatis (29.8%) and Dalits (19.5%). Overall, 53.8% of irrigation systems were damaged in one way or another. This proportion is particularly high in Sindhupalchok (87.8%), Rasuwa (76.3%), Ramechhap (76.3%) and Kabhrepalanchok (75%).
Among households who reported to have irrigation systems, those reporting damages are significantly more likely to expect reduced crops and to have a lower income.
Overall, 88.3% of the population surveyed reported to own livestock, although this varies across districts with Dhading, Okhaldhunga, Ramechhap and Sindhuli registering the highest share of livestock breeders (all over 94%). The animals most commonly bred are cattle (by 80% of households), followed by chickens, and sheep or goats (by 68 and 69% of households respectively).
Across the board, a majority of agricultural households (54%) do not report earning any income from agriculture, despite relying on it as part of their livelihoods for subsistence; only 27% report selling any crop and 22% sell either milk or meat. Crop sellers have significantly higher production than non-sellers of rice (nearly twice as much) and potatoes (over four times as much). Livestock product sellers (meat and milk), particularly poultry breeders, own twice as many animals on average (12) than non-sellers (6). These households suffered extensively from shelter collapse, as chickens were kept indoors at the time of structural collapse. 60% of sellers report decreased income from cereal sales and so do 53% and 34% of meat and milk sellers respectively. Households reporting damaged assets were found to be slightly more likely to report decreased income, however pre-existing patterns such as caste, the number of animals owned before the earthquake or crop production, as well as distance from markets are much more important determinants of agricultural product commercialization.
Access to Services
Access to services and resources, including secured water sources, improved sanitation facilities, health, education and finances was generally quite high but did exhibit variation across the assessed geographic areas. However, no significant relationship was found between levels of service access and socioeconomic characteristics. Nearly three in four (72.8%) households reported piped, municipal water as their primary drinking water source across all assessed areas, indicating steady supply and access to this public service, largely in line with pre-earthquake figures.2 The remaining households relied on a mix of protected (5.2%) and unprotected wells (5.7%), surface water, including natural springs and rivers (13.9%), and privately sourced bottled and trucked water (0.7%).
The use of surface water as a primary source of drinking water was nominally higher amongst rural households, 15% of whom sourced surface water for drinking, than amongst urban households, 8.2% of whom used this source. Similarly, use of surface water was more common amongst households in the high (16.4%) and mid hills (16.2%), in line with the general trend of reduced service provision in more hard-to-reach areas.
Access to basic sanitation infrastructure was much poorer. Overall, more than one in ten (12.6%) households had no access to latrines, indicating a high rate of open defecation. For instance, a quarter (24.5%) of households in Nuwakot, and over a fifth of households in Ramechhap (21.6%) and Rasuwa (21%) had no access to toilets.3
Access to health services was generally high, although again, different levels of access were observed across the assessed geographic areas. Overall, 15.5% of households reported experiencing constraints when attempting to access health services. The sex of the head of household was not found to be a determinant of service access, suggesting that access issues are primarily related to service supply rather than exclusionary practices. The district of residence of a given household seems to be a potent predictor of service access constraints, with 42.8% of households in Okhaldhunga, 33.4% of households in Rasuwa and 37.4% of households in Sindhupalchok reporting access constraints.
Access to education for children aged 5-16 was generally found to be high and at an estimated 95%, approaching universal enrolment across the assessed areas. It also largely aligns with pre-earthquake attendance rates, even exhibiting marginal increases in some districts. No variation, statistically significant or otherwise, was observed across elevations, the rural-urban divide or castes and social groups, indicating high service coverage, reach and participation.
Protection
Overall, 38.9% of surveyed individuals reported that they did not possess citizenship and/or identification documentation (including birth certificates) at the time of the assessment, potentially due to loss or damage as a result of the earthquake. There was little variation between districts. Sindhuli had the highest proportion of individuals (41%) who reported not being in possession of such documentation. Fewer individuals in rural areas (60%) had access to key documentation than in urban areas (69%). Possession of documentation also varied by caste/ethnicity: 54.6% of Dalit individuals possessed identification documentation compared to 68.2% of Brahmin/Chhetri individuals.
An estimated 86.9% of all households reported being in possession of land or property deeds. However, nearly a fifth (18.2%) of households in Rasuwa, 16.5% of households in Sindhuli and 15% of households in Gorkha were not in possession of land and/or property deeds at the time of the survey. The rural-urban divide was again a powerful determinant of possession: 14% of all households residing in rural areas were not in possession of deeds, whilst only 7.2% of urban households were not. Lack of such documentation may leave households more vulnerable to abuse and predatory practices, including arbitrary evictions, predatory rent pricing, or land grabs.