* This presentation explores a new flexible Bayesian approach for inference on income distributions using grouped data. Combining notions from Bayesian hierarchical modelling with previous results from the income distribution literature, a parametric 'missing rich'-expanded Generalized Lorenz curve model is developed in the interest of allowing for different assumptions and corrections on possible undercoverage and/or under-reporting of high incomes in the data. In fitting this model to the typical publicly-available grouped-data sources on household incomes, the unavailability of its likelihood function poses important limitations for Bayesian inference which can be overcome through an Approximate Bayesian Computation (ABC) approach. A particular adaptive ABC algorithm is proposed for the purpose of inference on the model's parameters and illustrated in both a simulated data example and a real data example on the French income distribution. Finally, ABC model choice methods are explored for the purpose of deciding amongst alternative candidate assumptions on the possible 'missing rich' dimensions of the data in the context of these illustrative examples.

** Since the financial crisis of 2008, many industrialized countries have suffered from low potential growth, low natural interest rates and low inflation. These three factors add up to an international phenomenon known as secular stagnation, to which central banks worldwide have reacted by adopting non-conventional monetary policies. In this theoretical paper, I evaluate the impact of monetary and fiscal policies on the real economy, using an endogenous growth model in which economic growth interacts with productive public expenditure. This is illustrated via Japan and the Eurozone. I find that both low and high balanced growth paths (BGPs) are stable when the money supply is exogenous and unstable when the money supply is endogenous. This result implies that a policy mix under quantitative easing has positive effects on real economic activity.