NSO Releases Model-Based District-Level Consumption Estimates for Uttar Pradesh

The report employs the Small Area Estimation (SAE) technique, a statistical approach widely used internationally to improve the precision of estimates for smaller administrative units.

The National Statistics Office (NSO), under the Ministry of Statistics and Programme Implementation (MoSPI), has released a pioneering report titled “Model-Based District-Level Estimates based on the Household Consumption Expenditure Survey (HCES) 2022–23” for the State of Uttar Pradesh. The study marks a significant step in India’s efforts to use advanced statistical modelling for generating district-level estimates of living standards—a move that can profoundly strengthen data-driven governance and localized policymaking.

The complete report is available on the MoSPI website: Model-Based District-Level Estimates Report – Uttar Pradesh.

A New Approach to Measuring Consumption at the District Level

Traditionally, the Household Consumption Expenditure Survey (HCES) has been one of the cornerstone surveys conducted by NSO, providing vital insights into consumption patterns, household characteristics, and living standards at national and state levels. However, due to smaller sample sizes at the district level, obtaining statistically robust consumption data for local administrative units has always been a challenge.

Recognizing this gap, the National Statistical Commission (NSC) Steering Committee on the National Sample Survey (NSS) recommended a pilot project to explore whether model-based estimation techniques could provide credible district-level statistics. Accordingly, a committee was established under the Chairpersonship of Dr. Mausumi Bose, Former Professor at the Indian Statistical Institute (ISI), Kolkata, to carry out the study using data from the HCES 2022–23.

The project was technically supported by both the NSO and the Directorate of Economics and Statistics (DES), Government of Uttar Pradesh.

Objective and Scope of the Study

The main objective of the pilot study was to develop district-level estimates of Monthly Per Capita Consumption Expenditure (MPCE) using statistical modelling methods that could supplement and improve upon direct survey results. By testing this approach in Uttar Pradesh, India’s most populous state, the study sought to establish whether small area estimation (SAE) models could bridge data gaps and produce reliable local-level insights for governance.

If successful, this methodology could be applied to other Indian states and indicators—such as poverty, health, education, and employment—thus enabling a more granular understanding of socio-economic progress across the country.

The Statistical Model: Small Area Estimation (SAE)

The report employs the Small Area Estimation (SAE) technique, a statistical approach widely used internationally to improve the precision of estimates for smaller administrative units. The method works by combining survey data with auxiliary administrative and demographic information, thereby “borrowing strength” from related data sources to enhance the reliability of district-level estimates.

For this study, two specific models were used:

1. Fay–Herriot (FH) Model

2. Spatial Fay–Herriot (SFH) Model

These models help capture spatial relationships between neighboring districts, ensuring smoother and more realistic estimates across regions.

Auxiliary Data Sources Used

To strengthen the model’s predictive accuracy, the study incorporated multiple administrative datasets representing socio-economic and welfare indicators. These included:

• Number of old-age pension beneficiaries

• Number of patients registered under the Ayushman Bharat (PM-JAY) health scheme

• Number of domestic LPG connections

• Number of households receiving benefits under the Antyodaya Anna Yojana (AAY)

By integrating such administrative data with the HCES dataset, the model effectively generated district-level MPCE estimates that are statistically stable and policy-relevant.

Key Findings: District-Wise Consumption Patterns

The analysis provides valuable insights into district-level variations in living standards across Uttar Pradesh—India’s most populous state with diverse socio-economic characteristics.

• Top Five Districts in Rural Areas (Highest Average MPCE):

  1. Bagpat

  2. Saharanpur

  3. Gautam Buddha Nagar

  4. Meerut

  5. Ghaziabad

• Top Five Districts in Urban Areas (Highest Average MPCE):

  1. Gautam Buddha Nagar

  2. Gonda

  3. Ghaziabad

  4. Bagpat

  5. Lucknow

These rankings reflect the economic diversity within Uttar Pradesh, where districts near the National Capital Region (NCR) exhibit higher consumption levels, while others demonstrate potential for targeted welfare and development interventions.

Significance of the Study

This pioneering effort demonstrates that model-based statistical techniques can serve as cost-effective, scalable tools for producing district-level data that would otherwise be unavailable due to survey limitations. It addresses a critical gap in India’s data infrastructure, enabling policymakers to design, monitor, and evaluate welfare schemes more effectively at the grassroots level.

The study also provides a proof of concept for leveraging statistical models to track socio-economic progress in alignment with the Sustainable Development Goals (SDGs) and India’s Viksit Bharat @ 2047 vision.

Future Applications and Policy Implications

The NSO report suggests that this model-based estimation framework can be expanded to cover other vital socio-economic indicators, including employment, poverty, education attainment, and health outcomes. By using reliable local data, district administrations, state planners, and central ministries can better identify regional disparities, allocate resources efficiently, and tailor development programmes to specific local needs.

Moreover, this approach supports the government’s emphasis on data-driven policymaking, as articulated in the National Data Governance Framework Policy (NDGFP). It underscores the potential of statistical innovation in improving transparency, accountability, and evidence-based decision-making across all tiers of governance.

Strengthening India’s Statistical Ecosystem

The success of this study reaffirms NSO’s crucial role in modernizing India’s official statistical system through the integration of traditional survey methods with advanced data science and modelling tools. It also highlights the importance of collaboration between national statistical agencies, state governments, and research institutions in generating robust, granular insights.

By pioneering this model-based district-level estimation, India is aligning itself with international best practices in official statistics and paving the way for more decentralized, data-empowered governance.

The release of the Model-Based District-Level Consumption Estimates for Uttar Pradesh is a landmark step in strengthening India’s statistical architecture. It showcases how innovative methods such as Small Area Estimation (SAE) can provide credible insights into local economies, helping policymakers better target poverty reduction, social protection, and development initiatives.

The report not only enhances understanding of consumption and living standards in Uttar Pradesh but also serves as a blueprint for future applications across other states and socio-economic domains. As India advances toward Viksit Bharat 2047, such data-driven approaches will be indispensable for ensuring inclusive, sustainable, and regionally balanced growth.