We propose an amplification mechanism of financial crises based on the information choice of investors. Adverse news about the solvency of a debtor raises the value of private information and therefore induces the acquisition of information. Informed investors rely more on private information...

Machine learning methods in asset pricing are often criticized for their black box nature. We study this issue by predicting corporate bond returns using interpretable machine learning on a high-dimensional bond characteristics dataset. We achieve state-of-the-art performance while maintaining...