The historical evolution of the wealth distribution: A quantitative-theoretic investigation

We derive qualitative and quantitative predictions of a microfounded model of wealth inequality and look at how the model's predictions compare with actual outcomes in the United States over the postwar period. The model's microeconomic core emphasizes that a household faces important earnings risks (deriving from the risk of unemployment and from uncertain wages) and cannot fully insure against these risks. To capture these features, we base our work on the broadly used Bewley-Huggett-Aiyagari (BHA) setup. Our model also features stochastic movements in saving rates that go beyond the mere precautionary-savings motive inherent in the BHA setup: we posit some randomness in discount factors, as in Krusell and Smith (1998). We demonstrate that such randomness gives rise to a Pareto-shaped right tail of the wealth distribution, a feature that appears to approximate the data well and that has been discussed much recently and derived in a variety of highly stylized models. Piketty and Zucman (2014) develop one such model and we argue that it can be viewed as a reduced form for the model we present here, and indeed also for the one in Krusell and Smith (1998).