Data project

Data Warehouse Labour Market and Social Protection (DWH LM&SP) – career data on retired individuals

Datawarehouse Arbeidsmarkt en Sociale Bescherming (DWH AM&SB) – loopbaangegevens gepensioneerden

Summary

Strengths The typical strengths associated with administrative data apply. The use of administrative data is cost-effective, data quality is high, non-response is inexistent, etc. Specifically for life-course researchers, there are several added advantages, such as lack of attrition in between waves, lack of memory bias, etc. Typical for the Belgian situation is that all administrative datasets contain the National Register number and can therefore accurately be linked. Weaknesses Apart from the typical weaknesses of administrative data (e.g. lack of data on opinions, motivations etc.), the following problems can be mentioned: (1) There is no information on educational levels; (2) Due to an evolution in register systems, the data contain some statistical breaks; (3) Data on personal and household characteristics only become available with a three year time lag; (4) Information on the Data Warehouse Labour Market and Social Protection is only available in French and Dutch; (5) Belgian social security is extremely complex. As a consequence, the data that follow from it are also highly technical. This means that it is almost impossible to use the data without a thorough and detailed knowledge of the Belgian social security system. We therefore strongly advise foreign researchers to collaborate with Belgian research teams that have experience with the data.

Type of data

Data Source
Registry

Type of Study
Other: longitudinal administrative data

Data gathering method
Registries

Access to data

Conditions of access
The data are available for academic and policy research. Researchers and policy-makers can either use the microdata or online modules to generate aggregate data.

Type of available data (e.g. anonymised microdata, aggregated tables, etc.)
Anonymised microdata; aggregated tables in online module

Formats available
SAS, Excel, PDF

Coverage

Coverage Years of collection, reference years, and sample sizes
1955-2012

First year of collection
Information on the careers of employees: 1955; Information on the careers of civil servants and self-employed: entire career for all retirees from 2001 onwards.

Stratification if applicable

Base used for sampling

Geographical coverage and breakdowns
Breakdown by region (Flanders, Wallonia, Brussels Capital Region), by province, by municipality and by district.

Age range
The total population is covered.

Statistical representativeness
Population representative

Coverage of main and cross-cutting topics
The DWH LM & SP is the most important dataset to study social systems and welfare in Belgium. To understand the data that are available, note that first pillar pensions in Belgium are calculated on the basis of the previous career (most importantly, the number of years worked and the wages earned). Hence, datasets have been set up that contain such career information. Separate datasets exist to calculate the pensions of employees, the self-employed and civil servants. The Sigedis dataset contains information on the careers of employees since the mid 1950’s. This career information is not only available for retirees, but also for individuals of active age. The career information used to calculate the pensions of the self-employed and civil servants, in contrast, is only available for retirees. The reason is that this information is only gathered at the time of retirement. The data on the self-employed are available from the National Institute for the Social Security of the Self-employed and the data on civil servants is from the Pension Service of the Public Sector `Pensioendienst voor de overheidssector / Service des pensions du secteur publique`.

Linkage

Standardisation
NACE-coding is used (i.e. statistical classification of economic activities in the European Community, developed by Eurostat)

Possibility of linkage among databases
The National Register number is integrated in all administrative datasets. In this way, information available in the Data Warehouse Labour Market and Social Protection can be linked to data from the National Register `Rijksregister / Registre National`, containing additional information on personal and household characteristics. As many other administrative datasets and survey datasets contain National Register numbers, it becomes possible to link the Data Warehouse Labour Market and Social Protection data to several other datasets. In this regard, it is important to notice that the Privacy Commission has stated under what conditions survey data from Statistics Belgium can be linked to the Data Warehouse Labour Market and Social Protection.

Data quality

Entry errors if applicable
General information on data quality of the DWH LM&SP: Data quality is high. However, given the detail of the information, data cleaning is far from straightforward and requires a thorough understanding of Belgian social security. Given the complexity of the recoding that is needed to make the data available for scientific research, errors can occur. These errors can be adjusted in collaboration with the Crossroads Bank for Social Security. Changes in legislation and registration can impact the content of the variables. Specific information related to the study of the careers of self-employed. The data on the self-employed are of lower quality than the data on employees and civil-servants.

Breaks

Consistency of terminology or coding used during collection

Governance

Contact information
Chris Brijs
Crossroads Bank for Social Security
Willebroekkaai 38
1000 Brussels Belgium Phone: +32 2 741 83 67
Email: chris.brijs(at)ksz-bcss.fgov.be
Url: www.ksz-bcss.fgov.be/nl/bcss/nodepage/content/websites/belgium/

Timeliness, transparency
The data are available two years following collection.