Income inequality and poverty indicators - Experimental statistics

Income inequality and poverty indicators

Experimental statistics logo

Why do we need flash estimates of income inequality and poverty indicators?

Providing timelier social statistics – especially indicators on income poverty and inequality – is a priority for the Commission and the European Statistical System.

Read moreLess

Indicators on poverty and income inequality are based on EU statistics on income and living conditions (EU-SILC). They represent an essential tool to prepare the European Semester (the annual cycle of economic policy coordination between EU countries).

Efforts for improving the timeliness of EU-SILC data are ongoing and significant progress has been done with the new Regulation 2019/1700.  However, the information on income collected in EU-SILC refers to the previous year and therefore monetary poverty and inequality indicators will always have a certain time lag (at least one year).

Flash estimates are calculated on the basis of a statistical or econometric model and have a release date appreciably earlier than the actual data. These estimates will complement the EU-SILC data and can be used in preliminary discussions and analysis until the final EU-SILC data become available.

Why are these indicators published as experimental statistics?

Their experimental nature is mainly related to the methodology used for their production which is based on microsimulation and macro-economic models. These methods are not traditionally used in the calculation of social statistics indicators.

As with any other estimate, the indicators should be interpreted with caution — their accuracy depends on several factors.  The flash estimates cannot perfectly capture changes in the EU-SILC estimates.

Although there are still limitations in the current methodology and its ability to replicate changes in EU-SILC, it can provide an early indication of the direction of change.

How are these indicators produced?

The key income indicators for which flash estimates will be available are:

  • AROP – at-risk-of-poverty rate for the total population
  • QSR – income quintile share ratio.

Read moreLess

To give a better overview of the evolution in income distribution, these are complemented by the changes in the nominal income deciles (D1, D3, MEDIAN, D7 and D9). These positional indicators are more sensitive to income changes and therefore suitable as early warnings.

As these indicators are experimental and in order to reflect the uncertainty involved in this kind of estimations, Eurostat decided to disseminate an uncertainty interval for the expected change, rather than point estimates.

The main approaches used are:

(1) Microsimulation

Microsimulation and nowcasting techniques are used to estimate the evolution of income poverty and inequality. This methodology takes into account both the evolution of employment incomes related to the labour market changes, as well as the impact of social policies. For the latter, estimations are based on EUROMOD, the tax benefit model at EU level.

(2) Flash estimates based on national sources

For a limited number of countries, flash estimates are based either on a) national registers data or b) data collected in national surveys via one or few questions on the current income. In the latter case, this differs from usual EU-SILC income indicators as they refer to the current reference period at the time of the interview (e.g. current month) while EU-SILC collects structural detailed information on income covering the whole previous year period.

Access the statistics

Feedback

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

  • Would you have comments or suggestions for improvements of the methods applied for this flash estimate exercise?
  • Are there any other factors Eurostat should consider?
  • What other indicators or breakdowns could be useful as early warnings on trends in income distribution and poverty?
  • Are there other indicators Eurostat should analyse for policy purposes?
  • Could the uncertainty interval be further improved? Would point estimates be desirable in the future?

Archive

2020 data

2019 data