The Causal Effects of Old Age Pensions

Value of consumption of food per capita in rupees over the last month. Households will be asked to report consumption of different food items over the last month in rupees; this will be added and divided by the number of household members to arrive at per capita values.

Yearly consumption on other goods and services in rupees. Households will be asked to report consumption of different goods and services over the last year in rupees; this will be added and divided by the number of household members to arrive at per capita values.

Household self-classification of its financial situation on 1-10 scale, with 1 being lowest and 10 being highest.

Whether any household member cut meals because of insufficient resources in the past year (0/1 variable)

Whether an elderly person cut meals because of insufficient resources in the past year (0/1 variable)

Participants will be asked over the last 30 days: how much difficulty they had in sitting for long periods, standing up from sitting down, getting up from lying down, standing for long periods, extending arms above shoulder level, washing whole body, getting dressed, getting to and using the toilet, taking care of household responsibilities, moving inside the home, walking 100 meters, walking long distances (1km), climbing a flight of stairs, stooping or kneeling, picking up things with fingers, carrying things, eating, getting where they want to go, concentrating on something for 10 minutes, learning a new task, and joining community activities. The scale is from 1-5, with 1 being no difficulty, 2 mild, 3 moderate, 4 severe, and 5 cannot do. The response for each activity will be converted to a z-score (subtract mean and divide by the standard deviation) and z-scores will be averaged to generate a composite index.

Participants will be asked to report severity of basic and acute health conditions over the last 30 days. Two outcomes: difficulty seeing and difficulty hearing someone are measured on a scale from 1-5, with 1 being no difficulty, 2 mild, 3 moderate, 4 severe, and 5 cannot do. Three outcomes: any problems with mouth or teeth in the last 12 months, discomfort urinating, and any falls or injuries in the last 12 months will be measured on a yes or no basis. The response for each activity will be converted to a z-score (subtract mean and divide by the standard deviation) and z-scores will be averaged to generate a composite index.

Participants will answer yes/no questions on whether they have any of the following health conditions: arthritis/rheumatism, stroke, chronic heart disease, hypertension and hypotension, diabetes, cancer, chronic lung disease, cholesterol, psychiatric illnesses, neurological disorders, kidney disease, liver disease. The response for each activity will be converted to a z-score (subtract mean and divide by the standard deviation) and z-scores will be averaged to generate a composite index.

Participants will be asked if they lost ~5kg without wanting to in the past 6 months (0/1 variable), gets enough food every day (0/1 variable), and number of meals per day. Each will be converted into a z-score and averaged.

The geriatric depression index short form is used. Participants are asked yes/no questions: if they are satisfied with their lives, have dropped activities and interests, feel their life is empty, often get bored, in good spirits most of the time, afraid that something bad is going to happen to them, feel happy most of the time, often feel hopeless, prefer to stay at home, rather than going out and doing new things, have problems with memory, think it is wonderful to be alive, feel worthless, feel full of energy, feel lonely. Responses will be averaged.

Cognition status, based on performance in cognition "quiz", i.e. Mini Mental State Examination (MMSE).

Participants will be asked if they fell ill in the last 30 days, and, if yes, whether they used healthcare to deal with the illness (0/1); They will also be asked how much money (in rs.) they spent on health expenditure over the last 30 days. This will be converted into an index by averaging z-scores.

UCLA Loneliness scale short-form

Whether cut meals in the last week (0/1), whether they lack resources to buy food in the following week (0/1), whether their family skipped meals (0/1), and whether there was a shortage of food in their household during the last week (0/1). Averaged into an Anderson index.

Whether they often felt lonely in the last week (0/1), were afraid bad things would happen to them (0/1), felt their situation was hopeless (0/1). Averaged into an Anderson index.

Whether they needed to but could not see a doctor in the last month (0/1), needed but could not get medication in the last month (0/1). Averaged into an Anderson index.

Whether they continued to receive financial support from their family (0/1), whether family continued contributing to household expenses (0/1). Averaged into an Anderson index.

Whether they continued to work during lockdown (0/1), whether they planned to resume work during lockdown (0/1), whether they had in-person visits (0/1). Averaged into an Anderson index.

Whether they got the COVID vaccine.

Whether they had someone take them to receive the COVID vaccine.

Whether still living at time of follow-up survey (0/1 variable)

Whether employed (0/1 variable)

Amount of time spent working (days/week, hours/day)

Amount of income earned in rupees

Number of adults in the household

Number of children in the household

Index of household asset ownership: car, two wheeler, TV, phone, refrigerator, washing machine, computer, air conditioner, internet.

Whether any member of household has outstanding debt or a loan (0/1 variable)

Whether elderly person has outstanding loan (0/1 variable)

Net amount of money sent (i.e. amount sent less amount received) to relatives outside of household in Rs.