Predictive regressions are subject to two small sample biases: the coefficient estimate is biased if the predictor is endogenous and asymptotic standard errors in the case of overlapping periods are biased downward. Both biases work in the direction of making t-ratios too large so that standard inference may indicate predictability even if none is present. Using annual returns since 1872 and monthly returns since 1927, the authors estimate empirical distributions by randomizing residuals in the vector autoregression representation of the variables. The estimated biases are large enough to affect inference in practice and should be accounted for when studying predictability.

Following its opening to trade and foreign investment in the mid-1980s, Mexico's economic growth has been modest at best, particularly in comparison with that of China. Comparing these countries and reviewing the literature, we conclude that the relation between openness and growth is not a simple one. Using standard trade theory, we find that Mexico has gained from trade, and by some measures, more so than China. We sketch out a theory in which developing countries can grow faster than the United States by reforming. As a country becomes richer, this sort of catch-up becomes more difficult. Absent continuing reforms, Chinese growth is likely to slow down sharply, perhaps leaving China at a level less than Mexico's real GDP per working-age person.

This article studies different methods of assigning a good to budget-constrained agents. Schemes that assign the good randomly and allow resale may outperform the competitive market in terms of Utilitarian efficiency. The socially optimal mechanism involves random assignment at a discount—an in-kind subsidy—and a cash incentive to discourage low-valuation individuals from claiming the good.

Predictive regressions are subject to two small sample biases: the coefficient estimate is biased if the predictor is endogenous, and asymptotic standard errors in the case of overlapping periods are biased downward. Both biases work in the direction of making t-ratios too large so that standard inference may indicate predictability even if none is present. Using annual returns since 1872 and monthly returns since 1927 we estimate empirical distributions by randomizing residuals in the VAR representation of the variables. The estimated biases are large enough to affect inference in practice, and should be accounted for when studying predictability.

Predictive regressions are subject to two small sample biases: the coefficient estimate is biased if the predictor is endogenous, and asymptotic standard errors in the case of overlapping periods are biased downward. Both biases work in the direction of making t ‐ratios too large so that standard inference may indicate predictability even if none is present. Using annual returns since 1872 and monthly returns since 1927 we estimate empirical distributions by randomizing residuals in the VAR representation of the variables. The estimated biases are large enough to affect inference in practice, and should be accounted for when studying predictability.

We apply machine-learning techniques to construct nonlinear nonparametric forecasting models of consumer credit risk. By combining customer transactions and credit bureau data from January 2005 to April 2009 for a sample of a major commercial bank’s customers, we are able to construct out-of-sample forecasts that significantly improve the classification rates of credit-card-holder delinquencies and defaults, with linear regression R2’s of forecasted/realized delinquencies of 85%. Using conservative assumptions for the costs and benefits of cutting credit lines based on machine-learning forecasts, we estimate the cost savings to range from 6% to 25% of total losses. Moreover, the time-series patterns of estimated delinquency rates from this model over the course of the recent financial crisis suggest that aggregated consumer credit-risk analytics may have important applications in forecasting systemic risk.

Traditional U.S. industries with higher firm-specific stock return and fundamentals performance heterogeneity use information technology (IT) more intensively and post faster productivity growth in the late 20th century. We argue that this mechanically reflects a wave of Schumpeter's creative destruction disrupting a wide swath of industries, with successful IT adopters unpredictably undermining established firms. This validates endogenous growth theory models of creative destruction and suggests intensified creative destruction as explaining findings associating greater firm-specific performance variation with higher per capita GDPs, economy growth rates, accounting standards, financial system development, and property right protection.

Regulators have expressed concerns about the “revolving door” between auditors and clients, whereby audit employees move directly from audit firms to audit clients (i.e., “direct alumni hires”). Regulators are concerned that these direct hires could compromise audit quality, partly because these employees could have previously audited their hiring company's financial statements. In contrast, we examine accounting and finance executives who move indirectly from audit firms to audit clients and who could not have previously audited the hiring company's financial statements (i.e., “indirect alumni hires”). We show that indirect hires occur more often than the direct hires that have concerned regulators. We predict and find that both direct and indirect alumni hires are associated with lower rates of executive turnover and audit firm turnover. However, there is no evidence that the reduced rates of executive turnover are explained by managerial entrenchment or that these hires are associated with lower audit quality. Overall, our findings suggest that direct and indirect employee movements from audit firms to audit clients are beneficial to executives, audit clients, and audit firms because they reduce the incidence of costly turnover.

We propose a methodology for studying changes in bilateral commodity trade due to goods not exported previously or exported only in small quantities. Using a panel of 1,900 country pairs, we find that increased trade of these “least-traded goods” is an important factor in trade growth. This extensive margin accounts for 10 percent of the growth in trade for NAFTA country pairs, for example, and 26 percent in trade between the United States and Chile, China, and Korea. Looking at country pairs with no major trade policy change or structural change, however, we find little change in the extensive margin.

We show that early joiners—nonfounder employees in the first year of a startup—play a critical role in shaping firm performance. We use administrative employer-employee matched data on U.S. startups and utilize premature death as a natural experiment that exogenously separates talent from startups. We find that losing an early joiner has large negative effects on employment and revenues that persist for at least ten years. In contrast, losing a later joiner yields only a small and temporary decline in firm performance. Our results imply that organization capital, an important driver of startup success, is embodied in early joiners.