Vortrag
Statistical Matching of Administrative and Survey Data: An Application to Wealth Inequality Analyses

Anika Rasner, Joachim R. Frick, Markus M. Grabka

2011 Joint Statistical Meetings : JSM 2011
Miami, USA, 30.07.2011 - 04.08.2011



Abstract:
Social security entitlements are a substantial source of wealth that grows in importance over the individual's lifecycle. Despite its quantitative relevance, social security wealth has been thus far omitted from wealth inequality analyses. In Germany, it is the lack of adequate micro data that accounts for this shortcoming. The two main contributions of this paper are: First, to elaborate a statistical matching approach that complements information on net worth as surveyed in the German Socio-Economic Panel (SOEP), a population representative panel study, with infor-mation on social security wealth from the Sample of Active Pension Accounts (SAPA), a large-scale administrative dataset maintained by the German Statutory Pension Insurance. Second, we show to what extent the inclusion of social security wealth affects the level and the distribution of individual net worth as well as overall inequality. The present value of pension entitlements (including entitlements from the statutory pension system as well as from the separate system for civil servants) amounts to 5.6 trillion Euros, which corresponds to an average of 78,500 Euros per person - thus almost doubling the level of net worth. Compared to results based on net worth only, inequality of our amended wealth measure is about 25 percent less. Moreover, we present significant differences in pension entitlements across occupational groups with civil servants gaining most from the inclusion of public pension wealth in the extended wealth meas-ure and self-employed benefiting the least. Overall, our results provide clear indication for the relevance of including the notional wealth held in pension entitlements providing a less biased picture of the level and the socio-economic structure of wealth in Germany. Above and beyond such within-country variation, our findings may also be most relevant for comparative analyses across welfare-regimes.