Theoretical life-cycle models are used to determine the optimal investment policy for a pension fund participant. These models are typically based on the assumption that individuals make rational choices. For such models, the risk aversion parameter, g (“gamma”), describes the amount of risk the individual is willing to take. The problem then is that the risk aversion we measure in experiments according to these models implies extremely risky investment policies and at the same time cannot explain the level of the equity premium. In addition, the rational model cannot explain why people save too little for their retirement and why there is high demand for a lump-sum payout in retirement in stead of an annuity.

This theoretical research introduces probability weighting into an otherwise standard preference model for life-cycle investing. This weighting provides subjective probabilities of survival and subjective expectations on financial returns. Based on the results in the literature, it seems that probability weighting within retirement savings can potentially explain three essential choices simultaneously that are considered anomalies under rational assumptions.