Premium
Credit Booms Gone Bust: Replication of Schularick and Taylor (AER 2012)
Author(s) -
Summers Peter M.
Publication year - 2017
Publication title -
journal of applied econometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.878
H-Index - 99
eISSN - 1099-1255
pISSN - 0883-7252
DOI - 10.1002/jae.2554
Subject(s) - econometrics , replicate , economics , bayesian probability , probit model , replication (statistics) , autoregressive model , bust , boom , computer science , statistics , artificial intelligence , mathematics , environmental engineering , engineering
Summary This paper replicates the results in Schularick and Taylor ( American Economic Review 2012; 102 (2): 1029–1061; ST hereafter). Specifically, I replicate ST's results in the ‘narrow’ sense by reproducing their calculations in the open source econometrics package gretl. (Gretl is an acronym for Gnu Regression, Econometrics and Time‐series Laboratory. It is available for Windows, Mac and Linux at www.gretl.sourceforge.net .) I also demonstrate the robustness of ST's findings to different estimation methods. I obtain qualitatively similar results to ST via Bayesian estimation of both static and dynamic panel probit models. Finally, I show that the marginal effects of credit growth on the probability of a financial crisis vary considerably across the countries in the dataset. Copyright © 2016 John Wiley & Sons, Ltd.