USAGE OF ADMINISTRATIVE DATA IN EU-SILC SURVEY

1. USAGE OF ADMINISTRATIVE DATA IN EU-SILC SURVEY Signe Bāliņa University of Latvia

2. EU-SILC • EU-SILC - European Union Statistics on Income and Living Conditions • EU-SILC is expected to become the EU reference source for comparative statistics on income distribution and social exclusion at European level

3. The Main Purpose and Main Tasks • The main purpose • To analyse the possibility of the usage of Latvian administrative registers for collecting income data necessary for EU-SILC survey • Main tasks • Gathering of information on existing administrative income registers and the analysis of them • The analysis of possible links and integration between data from EU-SILC survey and administrative registers

4. EU-SILC • EU-SILC • The reference population - all private households and their current members (16+) • Annual data: • Cross-sectional data pertaining to a given time or a certain time period with variables on income, poverty, social exclusion and other living conditions • Longitudinal data pertaining to individual-level changes over time, observed periodically over a four year period • Sample size in Latvia: • 3750 households per year • 7650 persons (16+)

5. EU-SILC Pilot • Population – all private households and their members (16+) • The method of sampling – a two-stage stratified random sample of households • Sample size • 500 households • The financial resources of the pilot survey allowed to survey 200 responding households (from 328, response level 61%) • In 200 responding households there were 505 persons from whom 408 persons (16+)

6. EU-SILC Pilot • The income data reference period in Latvian pilot survey was the preceding calendar year (year 2003), which for the respondents is a clear and unambiguous category • The information gathering method – direct interview • Household questioner • Person questioner

7. Identification of Persons and Related Problems • All permanent residents of Latvia have a unique person identification code • The person ID code can be effectively used as a key variable for merging different data bases • During the interviews of the 1st wave of the EU-SILC pilot survey the person ID code was not registered: it was suspected that registration of the person ID code may significantly decrease the survey response rate • After the 1st wave the person ID code was identified for 311persons or 60% of respondents • In the 2nd wave respondents were asked also about their person ID code: • 485 person ID codes were identified • For 20 persons identification of their person ID codes was impossible

8. Income Information of Administrative Registers • The main information sources: • State Revenue Service (SRS): • Monthly employers statutory declaration of income and tax about their employees • Statutory annual income declarations for self-employed persons (till April of the next calendar year) • The annual tax declaration of any tax payer (on voluntary base) • Specifying all income from different sources • Calculated and withhold taxes • Expenses redeemed from taxes (including contributions to the private pension funds) • State Social Insurance Agency information(SSIA) • Different type of pensions • Social benefits and allowances paid to Latvian residents

9. Information from SRS • Information requested about 485 person • Received income and tax information about 201 persons, from who were 178 respondents

10. Persons' Status at the Time of EU-SILC Pilot Study, no Information from SRS

11. Possible Information from SSIA • The SSIA register contains income information of several respondent groups of the EU-SILC survey: • pensioners – 30.5% of respondents • unemployed persons – 7.5% of respondents • persons receiving sickness benefits – 4.0% of respondents • persons receiving disability benefits – 3.3% of respondents • persons receiving maternity benefit – 2.5% of respondents • In the SSIA register it is possible to find some income information of more than 40% of all respondents of the EU-SILC pilot survey

12. Information about Wages and Salaries

13. Gross Wages and Salaries (Annual Data)

14. Gross Wages and Salaries, Annual Data (n=69) • Coefficient of correlation 0.983, p-value < 0.01 • SRS: • Averagewage Ls 2956.71 • Median Ls 1824.86 • EU-SILC • Averagewage Ls 2663.91 • Median Ls 1509.00

15. Gross Wages and Salaries, Monthly Data

16. Gross Wages and Salaries, Monthly Data (n=13) • Coefficient of correlation 0.139,= p-value =0.651 > 0.05

17. Net Wages and Salaries, Annual Data

18. Net Wages and Salaries, Annual Data (n=100) • Coefficient of correlation 0.884, p-value < 0.01 • SRS: • Average wage Ls 1944.98 • Median Ls 1211.21 • EU-SILC • Average wage Ls 2115.94 • Median Ls 1480.00

19. Net Wages and Salaries, Monthly Data (n=23) • Coefficient of correlation 0.864, p-value < 0.01

20. Conclusions • Inclusion of a person identification code in the EU-SILC questionnaire is necessary • It is no significant difference in annual gross/net wages and salaries from SRS and EU-SILC data • It is a significant difference in annual gross/net wages and salaries from SRS and EU-SILC data • It is necessary to solve the existing legislation problems in usage of the person level income data from SSIA register

21. Conclusions • The usage of register data: • Allows significant reduction of the response burden for persons participating in the EU-SILC survey • Allows obtaining detailed and more complete income data on wages and salaries • Promotes the survey response rate • Improves the total quality of the survey