Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/174417 
Year of Publication: 
2017
Series/Report no.: 
Working Paper No. 2017-08
Publisher: 
University of Massachusetts, Department of Economics, Amherst, MA
Abstract: 
In this paper we analyze Minskian dynamics in the US economy via an empirical application of Minsky's financing regime classifications to a panel of nonfinancial corporations. First, we map Minsky's definitions of hedge, speculative and Ponzi finance onto firm-level data to describe the evolution of Minskian regimes. We highlight striking growth in the share of Ponzi firms in the post-1970 US, concentrated among small corporations. This secular growth in the incidence of Ponzi firms is consistent with the possibility of a long wave of increasingly fragile finance in the US economy. Second, we explore the possibility of short-run Minskian dynamics at a business-cycle frequency. Using linear probability models relating firms' probability of being Ponzi to the aggregate output gap, which captures short-term macroeconomic fluctuations exogenous to individual firms, we find that aggregate downturns are correlated with an almost zero increased probability that firms are Ponzi. This result is corroborated by quantile regressions using a continuous measure of financial fragility, the interest coverage ratio, which identify almost zero effects of short-term fluctuations on financial fragility across the interest coverage distribution. Together, these results speak to an important question in the theoretical literature on financial fragility regarding the duration of Minskian cycles, and lend support, in particular, to the contention that Minskian dynamics may take the form of long waves, but do not operate at business cycle frequencies.
Subjects: 
Minsky cycles
financial fragility
firm behavior
JEL: 
B5
E32
G30
Document Type: 
Working Paper

Files in This Item:
File
Size
876.04 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.