We argue that dynamic indeterminacy in structural models can help rationalize statistical regularities regarding higher-order properties of macroeconomic time series. Without departing from the Gaussian rational expectations paradigm, we formally establish that any indeterminate equilibrium model admits a linear recursion with multiplicative noise representation. This allows self-fulfilling expectations (sunspots) to enhance endogenous propagation forces that trigger high-probability extreme changes in model variables, while also inducing time variation in conditional volatilities. As a result, even modest, short-lived exogenous shocks can produce large and persistent macroeconomic effects. Using a workhorse New Keynesian framework, we investigate the ability of such a general mechanism to account for observed fat-tailed behavior and volatility clusters in the inflation series over the Great Inflation period of US macroeconomic history.
Equilibrium indeterminacy and sunspot tales
Sorge, Marco M.
2021
Abstract
We argue that dynamic indeterminacy in structural models can help rationalize statistical regularities regarding higher-order properties of macroeconomic time series. Without departing from the Gaussian rational expectations paradigm, we formally establish that any indeterminate equilibrium model admits a linear recursion with multiplicative noise representation. This allows self-fulfilling expectations (sunspots) to enhance endogenous propagation forces that trigger high-probability extreme changes in model variables, while also inducing time variation in conditional volatilities. As a result, even modest, short-lived exogenous shocks can produce large and persistent macroeconomic effects. Using a workhorse New Keynesian framework, we investigate the ability of such a general mechanism to account for observed fat-tailed behavior and volatility clusters in the inflation series over the Great Inflation period of US macroeconomic history.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.