Income and consumption: social surveys and national accounts

Experimental statistics logo

Macro-level data on income and consumption from national accounts describe the situation of households as an institutional unit. In contrast, data on income distributions from social surveys, such as the EU Statistics on Income and Living conditions (EU-SILC) and data on consumption from the Household Budget Survey (HBS), are based on micro-level data and measure inequalities in the context of social policies.

Read moreLess

Due to differing concepts, definitions and data collection practices, these two data sources may lead to different conclusions as regards people’s prosperity. Thus, Eurostat proposes to reconcile data on income and consumption from social surveys and national accounts to obtain distributed household accounts. In an additional study, Eurostat proposes an approach to impute social transfers in kind from national accounts into EU-SILC micro data.

Reconciling social surveys with household accounts

By reconciling micro- and macro-economic data, Eurostat combines the high-level coherence of the integrated System of National Accounts with the distributional information of social statistics. These distributional accounts help to understand which households benefit the most from economic growth and which ones are more vulnerable to economic shocks.

Why publish these data as experimental statistics?

Comparing individual components of household income and consumption obtained from social surveys with national accounts reveals significant data gaps that need to be bridged. Uncertainty also arises from a lack of knowledge about the nature of part of these gaps, and thus the best method to fill them. This is particularly true for a reconciliation of micro and macro data at EU level, but even results produced at the national level by EU countries with access to additional information are affected by uncertainty. Hence, distributional national accounts data are treated as experimental. With their publication, Eurostat expects to collect useful feedback on the methodological choices and assumptions made in the production of these statistics, which will lead to their future improvement.

How are these statistics produced?

First, we compare social survey and national accounts household statistics and establish the best possible conceptual link between income and consumption items. Based onto this correspondence table, we calculate the value gap for each item. Then, we apply the following methods to allocate this gap and reconcile both statistics:

  • A proportional adjustment of the micro data to reach the value of the national accounts aggregate.
  • For certain income items, an adjustment of the top income micro data to follow a Pareto distribution; or allocation of shares of micro-macro gap by decile according to socio-economic assumptions.
  • After allocating the gap for each income and consumption item at household level, the total income and total consumption aggregates are rebuilt as the sum of available sub-items.
  • Households are then clustered into quintiles. For income, quintiles are defined according to the new total income aggregate; for consumption, according to the original survey income variable.

Read moreLess

For some countries (Greece, Lithuania, Luxembourg, Malta, Slovakia), we deviate from this approach applying a different combination of allocation methods by item on their request. Other countries (Czechia, France, Ireland, Netherlands, Slovenia, Sweden) provided their own estimates, which are based on more detailed information available at national level and applying methods that are best suited in the national context. These national data are published in a separate table. For further details, please consult the following documents:

Access the statistics

Imputing Social transfers in kind (STIK) into survey statistics

Social transfers in kind, which people receive from governments in form of free or subsidised education or health systems, cannot be collected through household surveys. This is because people are mostly unaware of the monetary value of such services. 

To account for this shortcoming, we propose a methodology for imputing social transfers in kind from national accounts into EU-SILC microdata. The aim is:

  • to obtain information on the distribution of STiK
  • to measure the impact of STiK onto the income distribution and on other measures of inequality, such as the Gini index.

Why publish these data as experimental statistics?

Information on social transfers in kind received by individual households cannot be obtained using surveys or administrative registers. Eurostat proposes the method described below as complementary information to official statistics.

How are these statistics produced?

Income distributions are obtained from EU-SILC microdata. Social transfers in kind are imputed using health profiles by age and gender from the European Commission’s ‘Ageing Report’ and are then reconciled with national accounts aggregates. For further details, please consult the following document:

Access the statistics


To help Eurostat improve these experimental statistics, users and researchers are kindly invited to give us their feedback:

  • Are statistics on distributional accounts for household income and consumption useful to you?
  • Are you interested into the distribution of social transfers in kind in micro data?
  • What’s your assessment of the methodology used? Do you have any ideas for further improvements?


2018 data comparison