Research‎ > ‎

Modeling status quo decisions: A cross-nested ordered probit model

ABSTRACT: The decisions to reduce, leave unchanged, or increase a choice variable (such 
as policy interest rates) are often characterized by abundant status quo outcomes 
that can be generated by different processes. The decreases and increases 
may also be driven by distinct decision-making paths. Neither standard nor 
zero-inflated models for ordinal responses adequately address these issues. This 
paper develops a flexible mixture model with endogenously switching regimes. 
Three latent regimes, which are interpreted in the interest rate setting context 
as loose, neutral and tight policy stances, create separate processes for rate 
hikes and cuts and overlap at a status quo outcome, generating three different 
types of zeros. The new model exhibits acceptable small-sample performances 
in Monte Carlo experiments, whereas traditional models deliver biased estimates. 
In the empirical application, the new model is not only highly favored 
by the statistical tests but also produces economically more meaningful inference 
with respect to existing models. More than one-third of the status quo
decisions are generated by the loose or tight policy stances, suggesting a high 
degree of intentional interest rate smoothing.

Download the latest versions of the paper and on-line appendix from the links below.