Household Income Survey: Getting responses to hard questions won't be easy

In an ideal world, the respondent to the Household Income Survey (HIS), to be undertaken by the National Statistics Office next year, would be like Vijay from the movie Deewar. Recall the immortalised dialogue, where Amitabh Bachchan (Vijay) tells Shashi Kapoor (Ravi): “Aaj mere paas bangla hai, gaadi hai, bank balance hai.” Granted, this is a conversation between two brothers and not between a surveyor and a respondent. If households are as forthcoming in their responses, conducting the HIS will be a breeze. 

In the past 75 years, the statistical system has undertaken surveys of consumption expenditure for two reasons. First, households are more comfortable disclosing consumption expenditure rather than income. Second, India had a food-security and hunger problem and it made sense to focus on consumption. This is not to suggest India does not collect information on the income of households. The annual Periodic Labour Force Survey has information on the labour-market earnings of all working members of the households. The Situation Assessment Survey (SAS) of Agricultural Households collects information from agricultural households on their various sources of income. Information on household income is available from other sources including the India Human Development Survey, Consumer Pyramid Household Survey, Nabard (National Bank for Agricultural and Rural Development) Financial Inclusion Survey, etc. 

As part of the planning for the HIS, the Ministry of Statistics and Programme Implementation has decided to exercise caution by conducting a pilot survey in order to understand the challenges in conducting the survey. The feedback from the field confirms the primary concern — 95 per cent of the respondents did not feel comfortable giving information on different sources of income to the interviewer. Since households tend to under-report their income, some will be classified as “Aamdani Atthanni Kharcha Rupaiya”. A majority of the respondents refused to answer questions on the income tax paid. For similar reasons, survey-based estimates of the number of households and individuals receiving benefits from various government programmes will always be lower than what is reported in the administrative data.

 

There is also a logistical challenge relating to the time of the day when the households should be visited. If the respondent is a working member, the person is busy getting ready for work in the morning, at work during the day, and tired in the evening. So can we have a non-working member of the household as a respondent? 

 

The data from the HIS is important because a majority of Indians work in the informal sector and do not file taxes. The survey is not only about collating information on labour income but also income from investment. It is entirely possible that the households may respond by saying they do not invest. This phenomenon, known as “satisficing”, occurs when respondents give what they deem is an okay response rather than an accurate response. This issue will crop up in the All India Debt and Investment Survey 2026. Households might be reluctant to give information on their financial assets. 

 

Stakeholders need to be sensitised on important issues. What if the response rate is well below 100 per cent as in current surveys? The non-response rate to questions on non-labour income is likely to be on the higher side. On the question of interest received in the last quarter, is it okay if respondents state that they do not know the answer. Or are approximate answers good enough? Do you, reader, know your interest income off the top of your head?

 

The timing of the Situation Assessment Survey of Agricultural Households will overlap with the HIS. Based on the Situation Assessment Survey, estimates of income of agricultural households are generated. Being a focused survey, the level of disaggregation at which each question is asked in this survey is higher than in the income survey. One insight from the survey-methodology literature is that when questions are asked at a high level of disaggregation, information is more accurately obtained, as compared to a single question. Unless analysts bear in mind such differences, a like-and-like comparison of estimates across two surveys will lead to erroneous conclusions.

 

If there are so many problems, then why even bother conducting the HIS? That is because we need to move the needle beyond estimates of poverty based on the survey of consumption expenditure and start a conversation on the adequacy of earnings. It is worth recalling that the terms “living wage” and “decent standard of life” find a mention in Article 43 of the Constitution of India.

 

Let’s acknowledge that executing the HIS 2026 is a non-trivial exercise. Given the challenges, the motto of the survey team should echo Circuit’s (Arshad Warsi) dialogue in the movie Munna Bhai MBBS “Full confidence mein janeka aur ekdum vinamra ke saath baat karneka”. 

 

The authors are professors at the Indira Gandhi Institute of Development Research, Mumbai, and Indian Statistical Institute, New Delhi, respectively.