Local Complementarities and Aggregate Fluctuations

Accumulating microeconomic evidence points both to firm level adjustment lumpiness and to the significant influence of idiosyncratic disturbances. Do these matter for aggregate fluctuations, or do their effects largely vanish upon aggregation? This paper explores the implications of local strategic complementarities and firmlevel adjustment lumpiness for aggregate fluctuations. It shows that small (industry-level), serially uncorrelated disturbances which are independent across industries can generate large, persistent aggregate fluctuations, even in the absence of aggregate shocks. This general amplification and propagation mechanism is explored in the context of a simple dynamic general equilibrium model. In this model, economy-wide fluctuations are driven entirely by independent industry-specific disturbances which are propagated via output market interactions. Results are encouraging: many key qualitative features of macroeconomic fluctuations are captured even by this simple model, indicating that this class of models, which builds upon microeconomic features that characterize the economy, surely merits further investigation. _______________________ Financial Support from the National Science Foundation is gratefully acknowledged. I would like to thank Donald Brown, Steven Durlauf, Chad Jones, Paul Milgrom, Uday Rajan, John Taylor, the Stanford macro lunch group, participants in workshops at Stanford University, Tufts University, Virginia Tech, and the 1995 NBER Summer Institute, and especially Robert Hall, Michael T. K. Horvath, and Ken Judd, for helpful criticism and suggestions. I also thank the Santa Fe Institute for its role in the evolution of this line of research and for its gracious hospitality in the 1992 Complex Systems Summer School, where I initially began thinking about this model, and to which some of the credit (if any) should go. All errors, misinterpretations, and omissions are mine. LOCAL COMPLEMENTARITIES AND AGGREGATE FLUCTUATIONS 2