Learning By Investing: Embodied Technology and Business Cycles

Abstract

In the last decade of the 20th century, the U.S. economy witnessed a persistent and substantial increase in private investment. The boom was sharply reversed in 2001, and a great deal of evidence suggests that the capital stock had become excessive. Standard equilibrium business cycle models have difficulties in predicting the investment boom and overshooting. An embodied technology model is constructed to replicate the pattern of investment boom and collapse. Unlike previous models of embodiment, the present model assumes that new technology increases the productivity of capital of all vintages, but only new capital can facilitate the adoption of the new technology. Further, although agents in this model know about the advent of a new technology, they have imperfect information about its magnitude. Agents learn the magnitude by investing in new capital. I present a sufficient condition for having a persistent investment boom and overshooting. I also solve the model numerically in a dynamic general equilibrium (DGE) setup. The model presented in this paper extends the standard DGE business cycle models in two ways: First, it presents a strong internal propagation mechanism with respect to technology shocks; second, it generates endogenous recessions without invoking technological regress. The model also offers a possible explanation on why consumption growth was strong during the last recession.