Projecting a Life-cycle Income – a Simulation Model for the Slovak Pension Benefit Statement

The introduction of a regulation requiring pension asset managers to provide savers with an estimation of pension benefits opened a wide range of scientific questions on the projection methods and estimation of input parameters. One of them is the estimation of life-cycle income for calculating expected contributions and the estimation of the benefit ratio at the moment of retirement. We present an estimation of life-cycle income functions for various age and educational cohorts influenced by temporary labor market shocks. By employing the resampling simulation method for incorporating macroeconomic shocks, we have shown that using longitudinal data on the income process from a large closed economy could bring valid results for a country with a small open economy as well where the longitudinal data on income processes of individuals are unavailable. Our findings could serve a practical use when pension or other social benefits tied to individual income should be modelled.

Higher-order income dynamics with linked regression trees

Summary We propose a novel method for modelling income processes using machine learning. Our method links age-specific regression trees, and returns a discrete state process, which can easily be included in consumption-saving models without further discretizations. A central advantage of our approach is that it does not rely on any parametric assumptions, and because we build on existing machine learning tools it is furthermore easy to apply in practice. Using a 30-year panel of Danish males, we document rich higher-order income dynamics, including substantial skewness and high kurtosis of income levels and growth rates. We also find important changes in income risk over the life-cycle and the income distribution. Our estimated process matches these dynamics closely. Using a consumption-saving model, the implied welfare cost of income risk is more than 10% of income.