Abstract This paper studies the impact of career concerns on technological change by analyzing the adoption of digital cinematography in the U.S. motion picture industry. This setting allows us to collect rich data on the adoption of this new technology at the project level (i.e., movie) and on the career of the main decision-maker (i.e., director). We find that early-career directors played a leading role in the adoption of digital technology, an effect that appears to be explained by career concerns, rather than alternative motives we consider and analyze. Technological savviness also plays a role. (JEL: G30, O33, L82, M50)Received: June 25, 2025Editor: J. Anthony Cookson Authors have furnished an Internet Appendix, which is available on the Oxford University Press Web site next to the link to the final published paper online.

Abstract We study how financial certifier competition influences loan contracting in the context of financial auditing. Exploiting the unexpected demise of Arthur Andersen that exogenously decreased auditor competition, we find a greater decrease in loan spread for borrowers in markets in which certifier competition declined more. Additional analyses suggest the result stems from enhanced audit quality and reduced credit risk. The effect of certifier competition is stronger for borrowers with weaker external monitoring and those generating significant revenue for their auditors. Our evidence highlights negative consequences of financial certifier competition. (JEL D43, G21, M42, M49)

Abstract This study examines the determinants and implications of the volume of tax-related numbers reported in the financial statements. I document that the volume of tax numbers increases with tax reporting requirements and decreases with the complexity of the tax rate, implying that firms with greater proprietary costs decrease their numeric disclosures once they meet mandatory reporting requirements. With respect to implications, I document that firms reporting more tax numbers improve the transparency of the information environment, reducing the errors and dispersion of analysts’ implied effective tax rate forecasts. In contrast, greater emphasis on narrative tax disclosure does not reduce information frictions, highlighting an important trade-off between numeric detail and strategic narratives. Further investigation suggests that this relationship is driven by more tax numbers in the financial statement footnotes rather than in the face financial statements. These findings suggest firms’ tax information environment improves with a greater volume of numeric tax-related disclosures.

Abstract We examine whether information processing constraints limit managers’ ability to effectively integrate tax planning and core business strategies (i.e., effective tax planning). We propose that artificial intelligence (AI) tools, such as machine learning, can mitigate these constraints by providing enhanced predictive information for key business decisions (e.g., customer demand, supply chain), thereby reducing processing costs. Using a recently developed firm-year measure of investment in AI-related human capital for a broad sample of U.S. nontechnology firms between 2010 and 2018, we find that AI investment is positively associated with tax effectiveness. This effect is concentrated among more complex firms and those where the tax function holds a higher status. Consistent with AI reducing information processing costs, we find that it improves tax effectiveness by enhancing internal information quality and internal capital management. We provide novel evidence that processing constraints hinder effective tax planning and show that AI can mitigate these constraints.

Abstract We study the role of public risk-sharing in markets where firms face substantial cost uncertainty, focusing on public reinsurance in health insurance. We develop a model where insurers internalize cost uncertainty through risk charges that raise effective marginal costs and create a role for reinsurance. Public reinsurance lowers both expected costs and cost volatility, particularly for smaller insurers, reducing prices and enhancing competition. Using an event study of staggered state-level reinsurance programs, we show that public reinsurance leads insurers to lower prices and private reinsurance purchases, benefiting financially constrained insurers the most. Structural estimates indicate that risk charges account for a substantial share of the premium-cost wedge and highlight public reinsurance’s comparative advantage over premium subsidies by providing risk protection and enhancing competition. Our results underscore the importance of accounting for firms’ risk exposure in policy design and provide a general framework for understanding public risk-sharing policies.

The long-term decline in US workers’ attempts to organize labor unions accelerated after 2000. We find that the swift rise of imports from China arising from a change in trade policy accounts for nearly all of this post-2000 acceleration: union certification elections decreased substantially among workers in manufacturing industries directly exposed to imports, but also among workers indirectly exposed through their local labor market. Consistent with a simple model of workers’ decision to seek union representation, direct exposure lowered the expected wage gain from unionization, whereas indirect exposure increased the cost of job loss—both of which discourage organizing.

Abstract Human decision-makers frequently override the recommendations generated by predictive algorithms, but it is unclear whether these discretionary overrides add valuable private information or reintroduce human biases and mistakes. We develop new quasi-experimental tools to measure the impact of human discretion over an algorithm on the accuracy of decisions, even when the outcome of interest is only selectively observed, in the context of bail decisions. We find that 90% of the judges in our setting underperform the algorithm when they make a discretionary override, with most making override decisions that are no better than random. Yet the remaining 10% of judges outperform the algorithm in terms of both accuracy and fairness when they make a discretionary override. We provide suggestive evidence on the behaviour underlying these differences in judge performance, showing that the high-performing judges are more likely to use relevant private information and are less likely to overreact to highly salient events compared to the low-performing judges.