We assess empirically whether monetary policy announcements impact firm expectations. Two features of our data set are key. First, we rely on a survey of production and price expectations of German firms, that is, expectations of actual price setters. Second, we observe the day on which firms submit their answers to the survey. We compare the responses of firms before and after monetary policy surprises and obtain two results. First, firm expectations respond to policy surprises. Second, the response becomes weaker as the surprise becomes bigger. A contractionary surprise of moderate size reduces firm expectations, while a moderate expansionary surprise raises them. Large surprises, both negative and positive, fail to alter expectations. Consistent with this result, we find that many of the ECB's announcements of non-conventional policies did not affect expectations significantly. Overall, our results are consistent with the notion that monetary policy surprises generate an information effect which is endogenous to the size of the policy surprise.

We assess how survey expectations impact production and pricing decisions on the basis of a large panel of German firms. We identify the causal effect of expectations by matching firms with the same fundamentals but different views about the future. The probability to raise (lower) production is 15 percentage points higher for optimistic (pessimistic) firms than for neutral firms. Optimistic firms are also more likely to raise prices. In a second step, we find optimism and pessimism to matter even if they turn out to be incorrect ex-post. Lastly, we quantify the contribution of incorrect optimism and pessimism to aggregate fluctuations.

No abstract