A Lot to Manage
Polarization, purpose and profit
Daniel Ferreira & Radoslawa Nikolowa
Journal of Financial Economics, October 2025
Abstract:
We present a model in which firms compete for workers who value nonpecuniary job attributes, such as purpose, sustainability, political stances, or working conditions. Firms adopt production technologies that enable them to offer jobs with varying levels of these desirable attributes. Firms’ profits are higher when they cater to workers with extreme preferences. In a competitive assignment equilibrium, firms become polarized and not only reflect but also amplify the polarized preferences of the general population. More polarized sectors exhibit higher profits, lower average wages, and a reduced labor share of value added. Sustainable investing amplifies firm polarization.
Job Mismatch and Early Career Success
Julie Berry Cullen, Gordon Dahl & Richard De Thorpe
NBER Working Paper, September 2025
Abstract:
How does being over- or underqualified at the beginning of a worker's career affect skill acquisition, retention, and promotion? Despite the importance of mismatch for the labor market, self-selection into jobs has made estimating these effects difficult. We overcome endogeneity concerns in the context of the US Air Force, which allocates new enlistees to over 130 different jobs based, in part, on test scores. Using these test scores, we create simulated job assignments based on factors outside of an individual's control: the available slots in upcoming training programs and the quality of other recruits entering at the same time. These factors create quasi-random variation in job assignment and hence how cognitively demanding an individual's job is relative to their own ability. We find that being overqualified for a job causes higher attrition, both during technical training and afterward when individuals are working in their assigned jobs. It also results in more behavioral problems, worse performance evaluations, and lower scores on general knowledge tests about the military taken by all workers. On the other hand, overqualification results in better performance relative to others in the same job: job-specific test scores rise both during technical training and while on the job, and these individuals are more likely to be promoted. Combined, these patterns suggest that overqualified individuals are less motivated, but still outperform others in their same job. Underqualification results in a polar opposite set of findings, suggesting these individuals are motivated to put forth more effort, but still struggle to compete when judged relative to others. Consistent with differential incentives, individuals who are overqualified are in jobs which are less valuable in terms of outside earnings potential, while the reverse is true for those who are underqualified.
Employee Performance and Mental Well-Being: The Mitigating Effects of Transformational Leadership During Crisis
Kristina Czura et al.
Management Science, forthcoming
Abstract:
The positive role of transformational leadership for productivity and mental well-being has long been established. Transformational leadership behavior may be particularly suited to navigate times of crisis that are characterized by high levels of complexity and uncertainty. We exploit quasi-random assignment of employees to managers and study the role of frontline managers’ leadership styles on employees’ performance, work style, and mental well-being in times of crisis. Using longitudinal administrative data and panel survey data from before and during the COVID-19 pandemic, we find that the benefits of different leadership styles depend on the environment: Employees of more transactional managers outperform those of more transformational leaders before the onset of the pandemic. During the pandemic, however, more transformational managers lead employees to better performance and mental well-being. We discuss potential explanations and implications.
Voice AI in Firms: A Natural Field Experiment on Automated Job Interviews
Brian Jabarian & Luca Henkel
University of Chicago Working Paper, September 2025
Abstract:
We study the impact of replacing human recruiters with AI voice agents to conduct job interviews. Partnering with a recruitment firm, we conducted a natural field experiment in which 70,000 applicants were randomly assigned to be interviewed by human recruiters, AI voice agents, or given a choice between the two. In all three conditions, human recruiters evaluated interviews and made hiring decisions based on applicants' performance in the interview and a standardized test. Contrary to the forecasts of professional recruiters, we find that AI-led interviews increase job offers by 12%, job starts by 18%, and 30-day retention by 17% among all applicants. To explain these results, we explore three channels. First, analyzing interview transcripts reveals that AI-led interviews elicit more hiring-relevant information from applicants compared to human-led interviews. Second, recruiters score the interview performance of AI-interviewed applicants higher, but place greater weight on standardized tests in their hiring decisions. Third, applicants accept job offers with a similar likelihood and rate interview, as well as recruiter quality, similarly in a customer experience survey. Moreover, when offered the choice, 78% of applicants choose the AI recruiter, and we find evidence that applicants with lower test scores are more likely to choose AI. Overall, we provide evidence that AI can match human recruiters in conducting job interviews while preserving applicants’ satisfaction and firm operations.
Cut to the Curve: Underrecognition and Talent Loss from Forced Ranking in a Multinational Firm
Brittany Bond
Management Science, forthcoming
Abstract:
This paper examines unintended consequences of enforcing a curve on performance rankings. I examine a multinational company where some employees are downgraded from the top levels of rankings due to a strict recognition cutoff. I show that downgraded employees are at least 34% more likely to voluntarily exit, and that attempts by the organization to manage employee concerns, particularly regarding concerns of fairness, envy, and self-image, do not have the intended retention effect. Underrecognized employees leave even though the organization avoids calibration bias, offers reassurance about their career prospects, and compensates them with larger bonuses than their top-ranked peers. In robustness checks, I show that under these conditions even high-performing employees not nominated to the top ranks are more likely to voluntarily depart despite receiving the largest bonuses. These findings suggest that, where underrecognition occurs due to the restriction of top rankings, the mechanisms producing demotivation are more powerful than the management strategies meant to combat them.
Artificial Intelligence in Team Dynamics: Who Gets Replaced and Why?
Xienan Cheng, Mustafa Dogan & Pinar Yildirim
NBER Working Paper, September 2025
Abstract:
This study investigates the effects of artificial intelligence (AI) adoption in organizations. We ask: (1) How should a principal optimally deploy limited AI resources to replace workers in a team? (2) In a sequential workflow, which workers face the highest risk of AI replacement? (3) How does substitution with AI affect both the replaced and non-replaced workers’ wages? We develop a sequential team production model in which a principal can use peer monitoring -- where each worker observes the effort of their predecessor -- to discipline team members. The principal may replace some workers with AI agents, whose actions are not subject to moral hazard. Our analysis yields four key results. First, the optimal AI strategy stochastically replaces workers rather than fixating on a single position. Second, the principal replaces workers at the beginning and at the end of the workflow, but does not replace the middle worker, since this worker is crucial for sustaining the flow of information obtained by peer monitoring. Third, the principal may optimally underutilize available AI capacity. Fourth, the optimal AI adoption increases average wages and reduces intra-team wage inequality.
Delegation to artificial intelligence can increase dishonest behaviour
Nils Köbis et al.
Nature, forthcoming
Abstract:
Although artificial intelligence enables productivity gains from delegating tasks to machines, it may facilitate the delegation of unethical behaviour. This risk is highly relevant amid the rapid rise of ‘agentic’ artificial intelligence systems. Here we demonstrate this risk by having human principals instruct machine agents to perform tasks with incentives to cheat. Requests for cheating increased when principals could induce machine dishonesty without telling the machine precisely what to do, through supervised learning or high-level goal setting. These effects held whether delegation was voluntary or mandatory. We also examined delegation via natural language to large language models. Although the cheating requests by principals were not always higher for machine agents than for human agents, compliance diverged sharply: machines were far more likely than human agents to carry out fully unethical instructions. This compliance could be curbed, but usually not eliminated, with the injection of prohibitive, task-specific guardrails. Our results highlight ethical risks in the context of increasingly accessible and powerful machine delegation, and suggest design and policy strategies to mitigate them.
Automation-Induced Innovation Shift
Lin William Cong et al.
NBER Working Paper, September 2025
Abstract:
We study how exposure to automation affects the nature and level of corporate innovation, which informs how innovation begets innovation. We document that firms with high robot exposure alter their technological focus over time and shift innovative activities towards AI which automation naturally complements through data accumulation. The shift is more pronounced for firms with greater data generation or prior AI-related research experience. Because AI patents are more costly (e.g., in labor input), albeit more general and original, firms with high automation experience a significant rise in R&D expenditure but an initial drop in patent quantity, before benefiting -- an innovation “J-Curve.” Our findings not only resolve the puzzle that globally firms invent less despite the greater research effort amidst rising automation, but also provides insights on the heterogeneous paths of innovation, all of which we rationalize in a parsimonious dynamic equilibrium model.
Corporate Hierarchy
Michael Ewens & Xavier Giroud
NBER Working Paper, August 2025
Abstract:
We introduce a novel measure of corporate hierarchies for over 2,500 U.S. public firms. This measure is obtained from online resumes of 16 million employees and a network estimation technique that allows us to identify hierarchical layers. Equipped with this measure, we document several facts about corporate hierarchies. Firms have on average ten hierarchical layers and a pyramidal organizational structure. More hierarchical firms have a more educated workforce, higher internal promotion rates, and longer employee tenure. Their operating performance is higher, but they face higher administrative costs. They are more active acquirers and produce more patents, but not higher-quality patents. They exhibit lower stock return volatility and more stable cash flows. We also examine how companies adjust their hierarchies in response to demand and knowledge shocks. We find that biotech companies increased their number of layers following the Covid-19 pandemic, while companies flattened their hierarchies following the adoption of artificial intelligence (AI) technologies. These findings are consistent with the theoretical predictions of existing models of corporate hierarchies.