Introduction

The disruptions in food supply chains and the economic crisis triggered by the Covid-19 pandemic brought millions into food-insecure conditions and indebtedness. A 2020 report of the High Level Panel of Experts on Food Security and Nutrition of the UN Committee on World Food Security says that between 83 and 132 million additional people experienced food insecurity because of the pandemic (HLPE 2020). The situation worsened because of disruptions in food supply chains, loss of livelihoods and earnings, and fluctuations in food prices (ibid.). The report of a survey conducted by the Centre for Sustainable Employment (CSE), Azim Premji University, in April and May 2020 captures the economic distress caused by the pandemic among rural people: 73 per cent of rural households reported reduced food consumption and many rural households reported increased indebtedness because of loss of employment during the lockdown (CSE 2020). A resurvey done by CSE in September–November 2020 showed that 59 per cent of rural households consumed less than they did before they lockdown (Nath, Nelson Mandela, and Gawali 2021).

This note examines the impact of the Covid-19 pandemic on food security and indebtedness among rural households. It is based on a telephone survey of 164 households from 26 villages across 13 States. The survey was conducted by the Pandemic Studies Unit of the Foundation for Agrarian Studies (FAS), Bangalore, in October 2020.¹ Of these 26 villages included in the 2020 telephone survey, 24 are part of the socio-economic surveys of villages conducted under the Project on Agrarian Relations in India (PARI) of the FAS in the past decade.² An advantage of the PARI database is that previous household surveys provide an understanding of the social and economic characteristics of the village economy. The basic socio-economic characteristics of the surveyed villages and households are given in Appendix Tables 1 and 2. The FAS conducted a rapid assessment survey of 52 households from 21 villages across 10 States of India in April 2020 through telephone interviews. Based on the evidence from this survey, Modak, Baksi, and Johnson (2020) found that many rural households faced increased indebtedness because of low levels of income and limited income support. This note discusses the impact of the Covid-19 pandemic on food consumption and the public distribution system (PDS), indebtedness and sale of assets, and income support in October 2020, six months after the previous survey.

Food Consumption

We asked the respondents about food consumption during the previous 30 days (September 2020), as compared to the situation before the lockdown. Forty-five per cent of the surveyed households (73 out of 164) consumed less food in the previous month than before the pandemic. These households were concentrated in the villages of Nayanagar and Katkuian (Bihar), Muhuripur (Tripura), and Warwat Khanderao (Maharashtra) (33 out of 73 households). In 11 of the 26 villages, 50 per cent or more of the interviewed households consumed less than they would have normally (Table 1). In eight villages – namely, Adat and Tholur (Kerala), Gharsondi (Madhya Pradesh), Bukkacherla (Andhra Pradesh), Kothapalle (Telangana), Alabujanahalli (Karnataka), and Mahatwar and Harevli (Uttar Pradesh) – no one reported consuming less food in September than they had in March.³

The proportion of households with less-than-usual food consumption was higher among Scheduled Caste and manual worker households as compared to households belonging to other castes and classes.⁴ Of 55 households belonging to Scheduled Castes or Scheduled Tribes, 31 (56 per cent) consumed less-than-usual food intake in September 2020. Among households belonging to other castes, the incidence of less-than-usual food consumption was 39 per cent (42 out of 109 households). This pattern was consistent across the villages. In five villages, all the surveyed Scheduled Caste households reported less-than-usual food consumption in September (Table 2).

It was mainly manual worker and poor peasant households that consumed less in September than in the pre-Covid period – 68 per cent (26 out of 38 households) of manual worker households and 38 per cent (17 out of 45 households) of poor peasant households.⁵ In eight villages, all respondents from manual worker households reported less-than-usual food consumption (Table 3).

We asked households about their food intake in the broad food categories of cereals, pulses, vegetables, fruits, meat, fish, eggs, milk, and sugar. A large proportion of households that had less-than-usual food consumption reported smaller intake of vegetables, milk, and meat than before.

Of 73 households with less-than-usual food intake, 10 reported that food consumption levels remained the same for children aged 6–14 but were lower for adults in the household during the Covid-19 period.

The common reasons for reduction in food consumption were reduced earnings, loss of employment opportunities, and a rise in prices of food commodities. Of the 24 sample households in Katkuian and Nayanagar villages in Bihar, 20 reported less-than-usual food consumption in September 2020. Many manual workers from these two villages are migrant workers in other States, and many of them had returned home in May because they lost their jobs. These villages also experienced heavy rainfall during the months of May and June, and crops were damaged as the fields were submerged. The crop failure affected the availability of food, mainly vegetables. In fact, in these two villages of Bihar, there were three rich/middle peasants also who reported less food consumption in September.

Public Distribution System (PDS)

The National Food Security Act (2013) classifies households into the following three categories for foodgrain entitlement under the PDS: households that are excluded with no entitlement (above the poverty line or APL), priority households (below the poverty line or BPL), and high priority or Antyodaya Anna Yojana (AAY) households (the poorest among the BPL households).⁶ We asked the respondent households about their PDS ration card and the items received in the reference month, that is, September 2020.

Of 73 households (64 per cent) reporting less-than-usual food consumption in the past 30 days, 46 had BPL cards and had received food items from the ration shop. The remaining 27 households (36 per cent) did not have BPL cards and had not received any food items from the ration shops. Respondents in Katkuian and Nayanagar (Bihar), Muhuripur (Tripura), and Amarsinghi (West Bengal) reported not receiving PDS food items because of problems related to their ration card.

Priority households (BPL cardholders) were entitled to 5 kg of foodgrain per person registered on the ration card or 25 kg of foodgrain per ration card for a month, whereas AAY households received 35 kg of foodgrain per card per month. There were variations in the food items received by households across States. In Andhra Pradesh, Bihar, Kerala, Maharashtra, Punjab, Tamil Nadu, and Tripura, households reported receiving pulses (chickpea, pigeon pea, red gram, lentils) along with foodgrains (rice or wheat). In other States, households received only foodgrains. A description of the items received by sample households is given in Appendix Table 3.

An additional allocation of 5 kg of foodgrain and 1 kg of pulses per person per month for BPL and AAY households was announced on March 26, 2020, under the Pradhan Mantri Garib Kalyan Anna Yojana, when the lockdown began in March (GoI 2020b).⁷ Among all 164 households, 81 had access to either BPL or AAY cards, and of these 81 households, only 39 reported receiving additional allocation of foodgrain and lentils during the pandemic.

There are regional variations in the access to PDS. Tamil Nadu follows universal PDS, so all the households in our study villages (Palakurichi and Venmani) had access to ration shops. A few households in Nayanagar (Bihar) and Muhuripur (Tripura) reported having AAY cards, which are for the poorest of poor households identified by the State Government. These households received 35 kg of foodgrain in the reference month (September 2020). Despite expansion of the PDS, several households faced difficulties in receiving foodgrain from ration shops. Case studies from Tamil Nadu and Tripura, cited below, illustrate the distinct regional scenarios of PDS provisioning.

AD is a manual wage worker from Tamil Nadu’s Palakurichi village (Nagapattinam district). His job is that of a contract worker at a circuit board manufacturing factory in Coimbatore. Between April and June 2020, when he had no work, his household members could manage only two meals in a day. His wife is a casual wage worker in the village, and they have two school-going children. After August, he returned to Coimbatore for work. Between September and October, his wife got 14 days of employment for weeding paddy, at a daily wage of Rs 200. The household received 25 kg of rice, 2 kg of pulses, 1.5 kg of sugar, and 1 litre of cooking oil from the PDS every month. AD’s two children received 5 kg of rice, 1 kg of pulses, and 10 eggs per month per child, in lieu of the mid-day meal they would have got in school. They managed to eat at least two meals a day because of access to PDS, despite an overall reduction in food intake.

ARD is a manual worker from Muhuripur village, South Tripura district, Tripura. He works as a wage labourer in rubber tree cutting, crop operations, and the Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS) in the village. During the lockdown, he had no earnings for three months, and his household could manage only two meals a day. They had access to a functioning PDS shop in the village, and in April–May 2020, they received 5 kg of foodgrain per person in the household, 2 kg of pulses, and 1 litre of cooking oil. In subsequent months the household did not receive any rations because the local panchayat office, which was supposed to prepare a list of poor households with BPL cards for the disbursal of food commodities, failed to do so. This, together with the soaring prices of food items in the market, meant that ARD was not able to procure sufficient food for the members of his household.

A government order dated July 31, 2020, stated that State Governments and Union Territories were to provide foodgrain and a food security allowance (equivalent to cooking costs) to children eligible under the Mid-Day Meal Scheme because of the closure of schools during the pandemic.⁸ Among the households we surveyed, 120 children of 77 households were eligible under this scheme. However, only 26 of these households received the allowance or provisions in kind, in lieu of the mid-day meal, during the lockdown. There was regular distribution of provisions in Ananthavaram (Andhra Pradesh), Adat and Tholur (Kerala), and Rewasi (Rajasthan). A description of the items provided is given in Appendix Table 4.⁹

Indebtedness

Rural households took loans from various sources to cope with the loss in employment and earnings. Of 164 households, 58 (35 per cent) availed loans after the lockdown, and of these, only 10 availed formal sector loans.¹⁰ Of the nine households in Palakurichi (Tamil Nadu) that took loans during the pandemic, seven took formal loans, borrowing mainly from microfinance institutions (MFIs) and by pledging gold in commercial banks. The 48 households that took informal loans borrowed from villagers, neighbours, local moneylenders, and friends and relatives.

Many households in the study villages of Bihar and Tamil Nadu had to borrow during the pandemic. In Nayanagar and Katkuian (Bihar), 15 out of 24 households (62 per cent) took loans, and in Palakurichi and Venmani (Tamil Nadu), 15 out of 18 households (83 per cent) took credit. Of the 73 households with less-than-usual food intake, 32 (43 per cent) took new loans to cope with the stress of the pandemic. Of these 32, 8 belonged to Nayanagar (Bihar), 6 to Katkuian (Bihar), and 4 to Palakurichi (Tamil Nadu). Thus, in the Bihar study villages, a total of 14 households with less-than-usual food consumption took informal loans to manage their expenses. In Palakurichi (Tamil Nadu), however, four out of five households with less-than-usual food consumption took formal loans to cope with the stress of the pandemic.

The distribution of borrowers by socio-economic class shows that manual workers and poor peasants (64 per cent) availed most of the loans. In four villages – Nayanagar (Bihar), Bukkacherla (Andhra Pradesh), Venmani (Tamil Nadu), and Harevli (Uttar Pradesh) – all manual worker households took loans during the pandemic, and in five villages, all poor peasants reported availing loans (Appendix Table 5).

Among Scheduled Caste households, 60 per cent (28 out of 47 households) borrowed from family and friends in the village. They were in immediate need of cash for living expenses and had barely any assets to rely on during such a crisis. All Scheduled Caste households that took loans held less than one acre of land, and eight of them were landless. Among the 109 households belonging to other social groups, only 29 borrowed, indicating that less than 30 per cent of the households in this category needed loans. In the Tamil Nadu villages, out of the 15 households that availed credit, 10 were Scheduled Caste households. The households sought all possible sources to get loans at the earliest, in order to manage their expenses and repay existing loans.¹¹ A respondent from Palakurichi (Tamil Nadu) said,

I pledged two sovereigns of gold for Rs 20,000. We needed money to run the house, as well as to repay MFI loans after the first month of lockdown. I took a loan of Rs 10,000 from a local moneylender, and I repay Rs 100 per day.

The nature of loans taken reveals that of the 58 borrower households, 14 mortgaged gold jewellery. Of these 14, 10 were Scheduled Caste households – four from Palakurichi (Tamil Nadu); three from Venmani (Tamil Nadu); and one each from Ananthavaram (Andhra Pradesh), Kothapalle (Telangana), and Adat (Kerala). Exemption from repayment of existing MFI loans and non-banking financial company loans was given only for the first two months of the lockdown; thereafter, households had to repay dues even in the absence of earnings and employment. This added to their burden, and they were forced to pledge their assets.

Sale of Assets

Of the 164 households surveyed, 26 (16 per cent) sold different types of assets during the pandemic. Scheduled Caste households were more likely to sell assets than other households. Of the 47 Scheduled Caste households, 10 (21 per cent) sold assets to cope with the crisis, whereas only 16 (14 per cent) of the remaining 109 households belonging to other social groups sold assets.

Of the assets sold by 26 households, livestock holdings were the most common, sold by 14 households; two households sold agricultural land, and one sold non-agricultural land. Table 4 shows the various assets sold by rural households in the study villages during the pandemic.

Two cases of land sale, from villages in Maharashtra and Punjab, are described below.

SB is a Scheduled Caste poor peasant from Warwat Khanderao village, Buldhana district, Maharashtra. He has a 1.5-acre landholding, cultivating cotton on one acre and soybean on the remaining land. He and other family members labour out during the cotton harvest season. Following the lockdown, they received meagre employment. Apart from reduced earnings, SB also faced difficulties in cultivation because of labour shortage and non-availability of inputs (seeds, fertilizers). He was forced to sell an acre of land in May 2020 as he was unable to deal with the rising input costs.

AS is a Scheduled Caste poor peasant from Hakamwala village, Mansa district, Punjab. He owns an acre of cultivable land, on which he cultivated wheat in December 2019. In April 2020, he harvested 12 quintals of wheat, of which he sold 10 quintals in the market at Rs 1,925 per quintal. His wife is an agricultural labourer who has received very little work since April 2020. She usually transplants paddy, but last year cultivators used machines to complete agricultural operations and therefore the household had reduced earnings. The price of inputs for cotton cultivation also rose significantly, and AS had to borrow Rs 40,000 from a moneylender in May 2020. He was forced to sell 0.012 acre of non-agricultural land for Rs 10,000 to pay the interest on the loan.

Livestock holdings are the primary assets that provide security to rural households during any form of economic crisis. Of the 10 Scheduled Caste households that sold assets in the surveyed villages, 6 sold livestock, whereas of the 16 households belonging to other castes that sold assets, 8 sold livestock. A Scheduled Caste respondent from Katkuian (Bihar) said,

I sold two goats during August for Rs 2,000 each. Before the lockdown, the price was Rs 2,500–3,000 each. I was forced to sell these because of no income during the lockdown.

Scheduled Caste households without landholdings sold poultry and goats at prices lower than in the pre-Covid-19 period to meet daily expenses in the absence of earnings and employment. There was distress sale of livestock holdings.

Income Support

As part of the fiscal stimulus announced by the Government of India on March 26, 2020, Rs 500 per month was to be provided to all women Jan-Dhan Yojana account holders between April and June 2020. Further, the first instalment of the Central Government’s Pradhan Mantri Kisan Sammann Nidhi (PM-KISAN) scheme, Rs 2,000, was to be paid in the first week of April 2020.

Of all 164 surveyed households, 71 (43 per cent) received income support from either State Governments or the Central Government. Of these, 34 reported support from PM-KISAN and 14 reported transfers under the Jan-Dhan scheme.

A disaggregation of households by caste shows that 21 out of 47 Scheduled Caste households and 49 out of 109 other caste households received income transfers. All Scheduled Caste households receiving any form of income support had less than one acre of landholding.

There was regional variation in the income transfer. In Rewasi (Rajasthan), Palakurichi and Venmani (Tamil Nadu), and Adat (Kerala), more than 80 per cent of households received income support, but no household in the two villages of Tripura reported receiving such support (Table 5).

Two case studies illustrate the differences in income support received by households.

SJ is a 36-year-old Scheduled Caste woman from Venmani village, Nagapattinam district, Tamil Nadu. She has neither an ownership nor an operational landholding. She is a manual wage worker engaged in coir and MGNREGS work. Her husband is a migrant construction worker, and she lives with her parents-in-law and two children. She and her mother-in-law together received only 47 days of work under MGNREGS from April to August 2020, and they jointly earned Rs 9,400. The reduced earnings forced SJ to pawn gold worth Rs 30,000 to repay the existing loans. She borrowed another Rs 15,000 from a moneylender in May to meet family expenses. The household received Rs 1,000 per month as assistance from the State Government against the ration cards of her family and parents-in-law. This amount helped them to survive the lockdown period. After September, the employment situation improved for the members of the household. They received 107 days of work, which included coir work, MGNREGS, and construction.

AD, a manual worker from Muhuripur village, South Tripura district, Tripura, is usually employed as a wage worker in agricultural operations and does MGNREGS work. Since April 2020, he has worked in rubber plantations and at transplanting paddy, less than 100 days in all. Despite the reduced earnings, he did not borrow money because he was unsure of his ability to repay loans; he sold livestock holdings instead. The household did not receive any assistance under State or Central schemes owing to the delay in preparation of the list of beneficiaries by the local administration.

While all households received Rs 1,000 in the two Tamil Nadu villages, only households with BPL cards received Rs 1,000 in Ananthavaram and Bukkacherla (Andhra Pradesh). An elderly Scheduled Caste woman respondent reported receiving Rs 18,750 as one-time aid for the year under Andhra Pradesh’s Cheyutha scheme (see Business Standard 2020). In the three West Bengal villages, a few households received income transfers from the State Government under the Krishak Bondhu Scheme (see Gupta 2020).

Conclusion

The Covid-19 pandemic affected the rural economy of India in several ways. In this note, we analysed its impact on food consumption, indebtedness, and coping strategies. The study is based on a survey of 164 rural households in 26 villages across 13 States. We found that, six months into the pandemic, around 45 per cent of the sample households reported a reduction in their usual food consumption. The impact was different for different socio-economic classes across the villages. The more vulnerable groups, for example, Scheduled Caste households and manual worker households, were worse affected. Around 68 per cent of manual wage workers and 38 per cent of poor peasants reported less food intake in September 2020 than in the pre-Covid period because of reduced employment and earnings and the soaring prices of food commodities. Fifty-six per cent of Scheduled Caste households reported lower food consumption that month than in the pre-pandemic period.

The public distribution system (PDS) played a pivotal role in providing food to households. Of the 73 households with less-than-usual food consumption, 46 had BPL cards. PDS functioning was more efficient in Andhra Pradesh, Punjab, and Tamil Nadu than in Bihar and Tripura. Several respondents in West Bengal and Tripura reported that they were unable to get food from the PDS. The reasons cited were bureaucratic delays and shortages of foodgrain stocks. Thirty-nine households received the additional allocation of foodgrain announced by the Central Government on March 26, 2020. Of 77 households with children aged 6–14 attending government schools, only 26 reported receiving foodgrain and food security allowance from the schools.

A little over a third (58 out of 164) of study households took loans to manage the livelihood crisis imposed by the pandemic. More than 80 per cent of these households took loans from informal lenders. In Katkuian and Nayanagar villages (Bihar), 70 per cent of households that reported less-than-usual food consumption in September 2020 had to take informal loans to meet their living expenses. Sixty per cent of all Scheduled Caste households reported taking loans, whereas the proportion for other caste households was 26 per cent.

Sixteen per cent of the households sold different types of assets. Scheduled Caste households (21 per cent) sold more assets than others (14 per cent). Distress sale of livestock holdings was prevalent.

A little over two-fifths of study households (43 per cent) received income transfers, from either the Central Government or State Governments. Of the 71 households that received income support, 48 per cent received it under the PM-KISAN scheme and 20 per cent under the Jan-Dhan scheme. Households received either Rs 500 (Jan-Dhan) per month from April to June or Rs 2,000 (PM-KISAN) as a first instalment in April 2020 – these amounts were, however, inadequate to compensate for the reduced earnings caused by the loss of employment during the pandemic.