Tatjana Xenia Puhan

We show that business cycle dynamics and, in particular, the cyclicality of a firm’s growth opportunities, determine the value of corporate cash holdings. An additional dollar of cash is more valuable for firms with less procyclical expansion opportunities. This valuation effect is strongest for low leverage and high R&D firms, but is independent of their financial status. Corporate cash holdings provide the flexibility to invest for firms that have expansion opportunities during crisis times with business cycle downturns and supply-side financial constraints. Cash holdings in firms with less procyclical growth opportunities are associated with higher investment and better operating performance.

This paper examines the predictive performance of machine learning methods in estimating the illiquidity of U.S. corporate bonds. We compare the predictive performance of machine learning-based estimators (linear regressions, tree-based models, and neural networks) to that of the most commonly used benchmark model based on historical illiquidity. Machine learning techniques outperform the historical illiquidity-based approach from both a statistical and an economic perspective. Moreover, tree-based models and neural networks outperform linear regressions, which incorporate the exact same set of covariates. Gradient boosted regression trees perform particularly well. While historical illiquidity, due to its high persistence, is the most important single predictor variable, several fundamental, risk-, and return-based covariates also possess notable predictive power. Capturing interactions and nonlinear effects among these predictors further enhances predictive performance.