Models of Sectoral Reallocation

This paper demonstrates several strengths and shortcomings of models of sectoral reallocation. Although such models demonstrate that sectoral reallocation can be an important amplification and propagation mechanism for exogenous shocks, they are essentially unable to explain any effects of sectoral reallocation on aggregate productivity or related quantities (such as the real wage or observations of aggregate increasing returns to scale), unless a wedge is introduced into the model that drives the marginal products of inputs in different sectors apart in steady state. In particular, costs of adjustment and lags to adjustment are not sufficient. This paper offers a solution to the problem in the form of variable sectoral capital utilization, the marginal product of which can differ across sectors in steady state. Reallocations of production between sectors in this setting are then shown to have first-order effects on aggregate productivity and real wages, and can explain the procyclicality of these variables without reliance on large, exogenous, and persistent shocks to technology