simulation model does not take into account socio-demographic processes, so the martial
status represents the status of the survey-year and any mortality is excluded.
The findings are based on the projected old-age-income that is calculated in the last step:
the gross old age incomes at the age of 65 are calculated on the basis of the individual
biographies, taking all relevant pension schemes - the SPI, the civil servants’ pension scheme,
the farmers’ old-age pension scheme, special schemes for the liberal professions, the public
and private supplementary systems and private provisions for old age (life insurances, private
pension insurances) - into account. Finally, the net incomes are generated by taking into
account current income tax regulations and a specially designed social insurance contribution
model.
The results of the AVID-Studies are published as reports and charts. The data have not
been published for the scientific community. However, for the combination of process
produced data from the pension fund’s record with survey data is the AVID the example for
further steps.
3. Future developments
3.1 Record-to-record linkage
Record-to-record linkage can improve the data, if other social security or tax institution
gathers information unknown to the pension fund. In the case of the Federal Employment
Agency this are data on training and other benefits from the unemployment insurance. In the
case of the public health insurance these would be information on prescriptions and health
treatments.
The record to record linkage is not an easy task for the research data centre of the pension
fund, because the social security number is not known for the persons included in the data
gathered for statistical purposes like the 500.000 persons included in the ‘Sample of the
insured population’ (VSKT). The public pension fund itself is composed of several regional
and two federal insurances, which keep the records of the persons insured by them. For
statistical purposes all pension insurers send data to the central statistical collecting point not
including the social security number. The research data centre is therefore unable to re-
identify the persons included in the data provided for scientific research. Record-to-record
linkage must therefore be supported from all pension insurers by sending the data including
the social security number for a particular research project and a particular year and data
source. These projects must therefore be thoroughly prepared in order to highlight the gain for
all participating parties and presented in the self-government boards. Strict data privacy rules
9