Yes, they can! Machine learning models predict leverage better than linear models and identify a broader set of leverage determinants. They boost the out-of-sample R2 from 36% to 56% over OLS and LASSO. The best performing model (random forests) selects market-to-book, industry median leverage, cash and equivalents, Z-Score, profitability, stock returns, and firm size as reliable predictors of market leverage. More precise target estimation yields a 10%–33% faster speed of adjustment and improves prediction of financing actions relative to linear models. Machine learning identifies uncertainty, cash flow, and macroeconomic considerations among primary drivers of leverage adjustments.

Abstract We show that the decision to go public is influenced by spatial variation in the supply of equity financing. We measure the amount of capital of equity investors in each U.S. region and document that the incidence of initial public offerings (IPOs) by intangible-intensive resident firms increases significantly when regional equity capital is abundant. Using a novel empirical strategy and hand-collected data on out-of-state pension flows, we confirm that our findings are not due to underlying regional factors.