Intersectoral Linkages, Diverse Information, and Aggregate Dynamics in a Neoclassical Model

What do firms learn from their interactions in markets, and what are the implications for aggregate dynamics? We address this question in a multi-sector real-business cycle model with a sparse input-output structure. In each sector, firms observe their own productivity, along with the prices of their inputs and the price of their output. We show that general equilibrium market-clearing conditions constrain average expectations and characterize a set of cases where average expectations, and therefore average dynamics, are exactly those of the full-information model. This "aggregate irrelevance" of information can occur even when sectoral expectations and dynamics are quite different under partial information, and despite the fact that each sector represents a non-negligible portion of the overall economy. Using sectoral data from the United States, we show that the conditions for aggregate irrelevance of information are far from being met in practice, yet the aggregate dynamics of the model calibrated to match US data remain nearly identical to the model under full-information.