Saving rates and portfolio choice with subsistence consumption

We analytically show that a common across rich/poor individuals Stone-Geary utility function with subsistence consumption in the context of a simple two-asset portfolio-choice model is capable of qualitatively and quantitatively explaining: (i) the higher saving rates of the rich, (ii) the higher fraction of personal wealth held in risky assets by the rich, and (iii) the higher volatility of consumption of the wealthier. On the contrary, time-variant 'keeping-up-with-the-Joneses' weighted average consumption which plays the role of moving benchmark subsistence consumption gives the same portfolio composition and saving rates across the rich and the poor, failing to reconcile the model with what micro data say.