Regime Changes and Real-Financial Cycles: Searching Minsky's Hypothesis in a Nonlinear Setting

TL;DR

This paper extends a nonlinear model to empirically test Minsky's financial cycle hypothesis across multiple countries, revealing regime changes and endogenous cycles linked to debt and interest rates from the 1970s to 2020.

Contribution

It introduces a nonlinear modeling approach to detect regime changes and Minsky cycles in real-financial data across several countries, expanding empirical validation.

Findings

Endogenous cycles found in all countries except Australia.

Interest rates are crucial for observing cycles across countries.

Interaction between household debt and GDP is significant only in the USA and UK.

Abstract

This paper investigates Minsky's cycles by extending the paper of stockhammer et al. (2019) with a nonlinear model to capture possible local real-financial endogenous cycles. We trace nonlinear regime changes and check the presence of Minsky cycles from the 1970s to 2020 for the USA, France, Germany, Canada, Australia, and the UK, linking the GDP with corporate debt, interest rate, and household debt. When considering corporate debt, the results reveal real-financial endogenous cycles in all countries, except Australia, and across all countries when interest rates are included. We find evidence for an interaction mechanism between household debt and GDP only for the USA and the UK. These findings underscore the importance of nonlinear regime transitions in empirically assessing Minsky's theory.