Working papers
Relocatable Capital: Theory and Empirical Evidence from 35 Million Lien Filings
This paper studies the dynamics of capital reallocation. Leveraging a novel dataset of over 25 million lien filings spanning four US states, this paper documents capital reallocation patterns for over two million unique pieces of heavy equipment and machinery at the serial number level. Reallocation of these capital goods is significantly procyclical and three times as volatile as aggregate output. A dynamic heterogeneous firm general equilibrium model with trade in used and new capital goods is developed to show that the coincidence of lower capital reallocation and lower output can be explained by the asymmetric nature of recessions: smaller, higher-growth firms are affected more in recessions than larger, lower-growth firms. This asymmetry implies that demand for used capital declines disproportionately more than the supply of used capital increases, which results in a lower quantity of used capital traded and a lower price of used capital during recessions. The implication of this explanation is that the lower rate of capital reallocation observed during recessions is partly due to an efficient response of the economy to asymmetric changes in the firm productivity distribution. The model is capable of rationalising a variety of investment-related business cycle moments including the puzzling negative correlation between productivity dispersion and capital reallocation observed in the data.
Work in progress
Machinery of Progress (with Peter Lambert)
In this paper, we chart the evolution of technological progress embodied in capital equipment used in production. We digitize archival legal filings of 10 million equipment purchases across four large US States going back to 1990. We link these transactions to firms and augment each record with a rich textual description of equipment capabilities using a novel methodology of Generated Natural Language Processing (GNLP). These data present a number of new facts about the adoption and diffusion of new equipment capabilities.
The Bankruptcy Accelerator
Does business bankruptcy and the resulting reallocation of capital from bankrupt to non-bankrupt firms attenuate or amplify aggregate fluctuations? I develop a dynamic general equilibrium model with endogenous default and show that the combination of bankruptcy and capital reallocation amplifies and propagates aggregate shocks. Following an aggregate decline in productivity, more firms choose bankruptcy and default on their loans. As a consequence, the cost of capital for bankruptcy-prone firms rises and more firms choose bankruptcy and default. This accelerator mechanism amplifies the initial shock by 23% in a model calibrated to the US economy and generates a much longer and deeper recession than a baseline model in which either the bankruptcy rate is kept fixed or in which capital is not allowed to reallocate.