Live on XaptiqUGC NET Economics (Dec 2026) and IIT JAM Economics (EN) test series are now live on Xaptiq.View test series
BusinessStatistical SystemStatisticsHousehold SurveyCPI Basket

MoSPI weighs diary-based HCES for gated societies to fix rising non-response

The Ministry of Statistics is considering a self-reported diary method for high-income households in gated societies for the 2027-28 Household Consumption Expenditure Survey, after urban non-response rose to 9.8% in 2022-23 from 2.8% in 2017-2018.

What happened

The Ministry of Statistics and Programme Implementation (MoSPI) is considering a diary-based data collection method for high-income households in gated societies for the next Household Consumption Expenditure Survey (HCES), expected to begin in mid-2027 and run for about a year. Under this method, households would enter their consumption details themselves, likely through an online portal, instead of being interviewed door-to-door. The move follows rising non-response from affluent urban households: the overall urban non-response rate jumped to 9.8% in the 2022-23 HCES from 2.8% in the 2017-2018 National Sample Survey.

Context

HCES is the survey used to measure how much households spend on different goods and services, and its expenditure shares feed into the Consumer Price Index basket. The most recent round, conducted from August 2023 to July 2024, covered 2.6 lakh households and asked about 405 items. After being run every five years historically, HCES has been placed on a three-year cycle. The non-response problem is largest among high-income urban households — 11% among the most affluent in the 2022-23 round, up from 3.3% in 2011-12 — and MoSPI held a brainstorming session with more than 75 housing societies in September 2024 to work out ways to raise participation.

Why it matters

HCES is the workhorse survey behind three of India's most important statistics: the CPI basket, poverty and inequality measures, and household consumption in the National Accounts. When non-response is concentrated in a specific income segment, the sample composition can drift from the intended one, biasing every downstream estimate — inflation weights, poverty headcount, distributional analysis. Fixing this matters because the RBI's monetary policy is anchored on a CPI target of 4% within a 2-6% band, which the CPI basket weights indirectly shape.

Impact for India

  • RBI: More representative CPI weights (built from HCES expenditure shares) would give the MPC a cleaner read of the retail inflation it targets.
  • Poverty and inequality estimates: Reducing under-representation of high-income households improves the top-of-distribution tail that drives inequality measures.
  • National Accounts: Household final consumption in GDP relies on HCES-based benchmarks; better coverage of gated societies improves the estimate.
  • Policy design: Data-driven welfare targeting depends on unbiased HCES data; skewed samples raise the risk of misallocated transfers.
  • Households: Diary-based reporting reduces recall error for the segment that opts in, but a two-method design creates a stitching challenge across income groups.

Key data

Households in most recent HCES, 2023-24: 2.6 lakhSurvey covered 405 goods and services
Urban non-response rate, 2022-23 HCES: 9.8%Up from 2.8% in the 2017-2018 NSS round
Rural non-response rate, 2022-23 HCES: 4.1%Up from 1.5% in the 2017-2018 round
Non-response among most affluent urban households, 2022-23: 11%Up from 3.3% in 2011-12
RBI CPI inflation target: 4%With a 2-6% tolerance band

Concepts

Household Consumption Expenditure Survey (HCES)MoSPI's sample survey of household spending on goods and services; used to measure standards of living, inequality and to update the CPI basket.The next round, expected from mid-2027, may adopt a diary-based method for high-income households in gated societies.
Non-response biasA distortion in survey estimates that arises when the households that refuse to respond differ systematically from those who do.Urban non-response rose to 9.8% in 2022-23, concentrated among affluent households.
Diary vs recall methodTwo collection modes for expenditure data: households log purchases as they occur (diary), or an enumerator asks them to remember past spending (recall). Diaries reduce recall error but require literacy and effort.MoSPI is weighing the diary method for gated societies where door-to-door surveys are increasingly refused.
CPI basket weightThe share assigned to each item in the Consumer Price Index, derived from household expenditure shares; heavier weights mean bigger price shocks matter more for headline inflation.HCES expenditure shares feed directly into the CPI basket the RBI targets.

Theory lens

Selection-bias and total-survey-error framework — total survey error decomposes into sampling error and non-sampling error, and rising non-response from a specific stratum inflates the non-sampling component in ways a bigger sample alone cannot fix. If the missing observations differ systematically from respondents, every downstream statistic — CPI weights, poverty headcount, Gini — is estimated on a sample that no longer matches the target population. The diary-based pilot is a mode-switch response, trading one bias (recall error in the respondent group) for the risk of a mode-mix bias between diary and door-to-door observations.

Stakeholders

Gains: Statistical agenciesDiary-based collection cuts the recall error and raises response rates in the segment that most refuses door-to-door surveys.
Gains: Users of CPI and poverty statisticsBetter sample representativeness reduces biases in inflation weights and inequality measures.

Practice Question — from the story

With reference to India's Household Consumption Expenditure Survey (HCES) and MoSPI's proposed diary-based method, consider the following statements:

  1. The urban non-response rate in the 2022-23 HCES rose to 9.8% from 2.8% recorded in the July 2017-June 2018 National Sample Survey round.
  2. The diary-based data collection method being considered would be extended to all households across urban and rural areas from the very first pilot.

Which of the statements given above is/are correct?

  1. 1 only
  2. 2 only
  3. Both 1 and 2
  4. Neither 1 nor 2
Reveal answer & explanation

Answer: 1 only

  • Statement 1: Correct. The article reports the overall urban non-response rate rose to 9.8% in 2022-23 from 2.8% in the 75th round of the NSS (July 2017 to June 2018).
  • Statement 2: Incorrect. The diary method is being considered only for high-income households in gated societies on a pilot basis before the 2027-28 HCES, not for all households.

Therefore, the correct answer is 1 only.

Concept check — test your understanding

In sample surveys such as the HCES, if the households that refuse to respond are systematically wealthier than those who do, the resulting estimates suffer primarily from which one of the following?

  1. Random sampling error, which shrinks as the sample size grows and can be handled by drawing more households.
  2. Recall bias, arising because respondents forget details of past expenditure.
  3. Non-response bias, a form of selection bias that a larger sample cannot fix if refusers systematically differ from respondents.
  4. Measurement error, arising because the survey instrument records the reported value incorrectly.
Reveal answer & explanation

Answer: C

  • Non-response bias arises when the missing observations differ systematically from the responding ones; simply drawing a larger sample from the same population does not correct it because the missing sub-population remains missing.
  • Option A describes random sampling error, which does shrink with sample size — non-response bias does not.
  • Option B is recall bias, a real concern for the diary/recall choice inside a household that does respond, but not the bias that arises from refusers.
  • Option D is measurement error — a separate class of non-sampling error tied to how the response is recorded, not to who refuses to respond.