Technology shocks and asset pricing: the role of consumer confidence

We show that the introduction in a power utility function of a confidence index to signal the state of the world allows for an otherwise standard asset pricing model to match the observed consumption growth volatility and excess returns with a reasonable level of relative risk aversion. Our results stem from two quantitative exercises: a calibration and a non-linear estimation. In both cases, our findings are robust to different data frequencies and various indicators of confidence. Our estimations are also robust to a number of instrument specifications. We rationalise this finding by developing a model where monopolistically competitive firms are subject to idiosyncratic shocks, which affect both the quantity and the quality of the goods produced. When households foresee good times, they expect firms to generate higher profits and produce higher quality goods. While greater expected excess returns provide a larger incentive to save, better expected quality of consumption discourages saving, as it lowers the expected marginal utility of any given level of physical consumption. Compared to standard consumption-based frameworks, our model thus predicts a more stable consumption path. Building on the customary notion of confidence indicators as the household expectations on the future state of the economy, we argue that confidence provides a suitable proxy for the unobservable quality of consumption via the positive correlation between the latter and the overall performance of the economy.