Firms' Optimism and Pessimism

Are firms' expectations biased? Does it matter? We use micro data on firms' production expectations from the German IFO Business Climate Survey and compare them to realization data from the same survey. We then construct series of quantitative firm-specific expectation errors. We find that depending on the exact definition at least 6 percent and at most 35 percent of firms consistently over- or underpredict their one-quarter-ahead upcoming production. In a simple frictionless neoclassical heterogeneous firm model these expectational biases lead to factor misallocations that cause welfare losses that in the worst case are comparable to conventional estimates of the welfare costs of business cycles fluctuations. In more conservative calibrations the welfare losses are even smaller.