ABSTRACT We use machine learning methods and high‐dimensional detailed financial data to predict the direction of one‐year‐ahead earnings changes. Our models show significant out‐of‐sample predictive power: the area under the receiver operating characteristics curve ranges from 67.52% to 68.66%, significantly higher than the 50% of a random guess. The annual size‐adjusted returns to hedge portfolios formed based on the prediction of our models range from 5.02% to 9.74%. Our models outperform two conventional models that use logistic regressions and small sets of accounting variables, and professional analysts’ forecasts. Analyses suggest that the outperformance relative to the conventional models stems from both nonlinear predictor interactions missed by regressions and the use of more detailed financial data by machine learning.

We use data from a large survey of Chinese firms to investigate whether local government efforts to facilitate market development improve firm efficiency. Both government provision of information about products, markets, and innovation and government assistance in arranging loans are positively associated with firm efficiency, and those private firms with weak access to and knowledge of financial, input, and product markets benefit most from such assistance. These patterns are robust across multiple estimation approaches. Our examination of the determinants of local government facilitation also suggests that it gravitates toward promoting efficiency, though there are also indications that rent-seeking may play a role. Our evidence is consistent with the notion that government facilitation can help some firms overcome market failures in the early stages of a country's private sector development. Though causality is difficult to establish, we argue that changing fiscal dynamics that forced local governments to become increasingly self-reliant in generating revenue, and a government promotion system based on local economic performance, were key motivating factors for market facilitation by local government officials.

This article provides new evidence on the housing-wealth effect on consumption using household panel data. A key advantage in studying the Chinese housing market is the absence of the collateral channel because households are prohibited from withdrawing housing equity. The results show that for every 1% increase in housing wealth, household consumption increases by 0.14%, suggesting an implied marginal propensity to consume out of housing wealth of 0.023. Further, we find that this marginal propensity to consume is the largest among employees who face greater income uncertainty, suggesting that precautionary-saving motives are driving the results.