Getting to Work
The Changing Role of Managers
American Journal of Sociology, September 2023, Pages 439-484
This study argues that the increase in middle management in recent decades was accompanied by a shift in managerial roles. Increased task complexity and a new management philosophy have reduced the need for direct supervision but generated a greater demand for collaboration, leading to the emergence of a managerial class whose primary role is collaboration not supervision. The author analyzed a large volume of data to generate three sets of findings: (1) The expectations of the managerial role have quickly changed, in almost all sectors, to emphasize more collaboration and less supervision (2) This new managerial role is especially concentrated in innovation-focused firms. (3) Firms treating managers as collaborators have a higher proportion of middle managers than those still treating them primarily as supervisors. These findings suggest that the role of managers has fundamentally shifted and that accounting for changing managerial roles could explain a significant portion of the managerial growth.
Network Referrals and Self-Presentation in the High-Tech Labor Market
Santiago Campero & Aleksandra (Olenka) Kacperczyk
Organization Science, forthcoming
The practice of recruiting job candidates sourced through social contacts (i.e., referrals) is pervasive in the labor market. One reason employers prefer to recruit through referrals is that these candidates often present resumes that are perceived to be a better fit for the role. Whereas existing research attributes this pattern to how individuals who make referrals (i.e., referrers) select individuals to refer, we propose a new mechanism: differences in self-presentation. We argue that referral ties increase the candidates' propensity to engage in self-presentation work, motivating and assisting candidates in presenting their backgrounds to convey fit. We examine this claim by utilizing unique data from an applicant-tracking system containing job applications for positions at U.S.-based high-tech firms between 2008 and 2012. A candidate fixed-effects specification reveals that when a candidate applies to a firm via a referral, he or she tends to showcase a rendition of his or her career history that better matches the target job than when the candidate pursues positions without such ties. Several mechanism checks, combined with supplementary survey evidence, further indicate that the presence of referral ties to the target firm is associated with greater motivation to engage in self-presentation work as well as the provision of different forms of assistance in that work.
Congestion in Onboarding Workers and Sticky R&D
Justin Bloesch & Jacob Weber
Federal Reserve Working Paper, November 2023
R&D investment spending exhibits a delayed and hump-shaped response to shocks. We show in a simple partial equilibrium model that rapidly adjusting R&D investment is costly if the probability of converting new hires into productive R&D workers ("onboarding") is decreasing in the number of new hires ("congestion"). Congestion thus causes R&D producing firms to slowly hire new workers in response to good shocks and hoard workers in response to bad shocks, providing a microfoundation for convex adjustment costs in R&D investment. Using novel, high-frequency productivity data on individual software developers collected from GitHub, a popular online collaboration platform, we provide quantitative evidence for such congestion. Calibrated to this evidence, a sticky-wage new Keynesian model with heterogeneous investment-producing firms subject to congestion in onboarding and no other frictions yields hump-shaped responses of R&D investment to monetary policy shocks.
Remote collaboration fuses fewer breakthrough ideas
Yiling Lin, Carl Benedikt Frey & Lingfei Wu
Nature, 30 November 2023, Pages 987-991
Theories of innovation emphasize the role of social networks and teams as facilitators of breakthrough discoveries. Around the world, scientists and inventors are more plentiful and interconnected today than ever before4. However, although there are more people making discoveries, and more ideas that can be reconfigured in new ways, research suggests that new ideas are getting harder to find -- contradicting recombinant growth theory. Here we shed light on this apparent puzzle. Analysing 20 million research articles and 4 million patent applications from across the globe over the past half-century, we begin by documenting the rise of remote collaboration across cities, underlining the growing interconnectedness of scientists and inventors globally. We further show that across all fields, periods and team sizes, researchers in these remote teams are consistently less likely to make breakthrough discoveries relative to their on-site counterparts. Creating a dataset that allows us to explore the division of labour in knowledge production within teams and across space, we find that among distributed team members, collaboration centres on late-stage, technical tasks involving more codified knowledge. Yet they are less likely to join forces in conceptual tasks -- such as conceiving new ideas and designing research -- when knowledge is tacit. We conclude that despite striking improvements in digital technology in recent years, remote teams are less likely to integrate the knowledge of their members to produce new, disruptive ideas.
Great Recession Babies: How Are Startups Shaped by Macro Conditions at Birth?
Daniel Bias & Alexander Ljungqvist
Vanderbilt University Working Paper, November 2023
We combine novel micro data with the quasi-random timing of patent decisions over the business cycle to estimate the effects of being born in the Great Recession on innovative startups. After purging ubiquitous selection biases and sorting effects, we find that recession startups experience better long-term outcomes in terms of employment and sales growth (both driven by lower mortality) and future inventiveness. While funding conditions cannot explain differences in outcomes, a labor market channel can: recession startups are better able to retain their founding inventors and build productive R&D teams around them.
Transmission Versus Truth, Imitation Versus Innovation: What Children Can Do That Large Language and Language-and-Vision Models Cannot (Yet)
Eunice Yiu, Eliza Kosoy & Alison Gopnik
Perspectives on Psychological Science, forthcoming
Much discussion about large language models and language-and-vision models has focused on whether these models are intelligent agents. We present an alternative perspective. First, we argue that these artificial intelligence (AI) models are cultural technologies that enhance cultural transmission and are efficient and powerful imitation engines. Second, we explore what AI models can tell us about imitation and innovation by testing whether they can be used to discover new tools and novel causal structures and contrasting their responses with those of human children. Our work serves as a first step in determining which particular representations and competences, as well as which kinds of knowledge or skills, can be derived from particular learning techniques and data. In particular, we explore which kinds of cognitive capacities can be enabled by statistical analysis of large-scale linguistic data. Critically, our findings suggest that machines may need more than large-scale language and image data to allow the kinds of innovation that a small child can produce.
Can a computer outfake a human?
Jane Phillips & Chet Robie
Personality and Individual Differences, February 2024
Faking on personality tests continues to be a challenge in hiring practices, and with the increased accessibility to free, generative AI large language models (LLM), the difference between human and algorithmic responses is difficult to distinguish. Four LLMs-GPT-3.5, Jasper, Google Bard, and GPT-4 were prompted to provide ideal responses to personality measures, specific to a provided job description. Responses collected from the LLM's were compared to a previously collected student population sample who were also directed to respond in a ideal fashion to the same job description. Overall, score comparisons indicate the superior performance of GPT-4 on both the single stimulus and forced-choice personality assessments and reinforce the need to consider more advanced options in preventing faking on personality assessments. Additionally, results from this study indicate the need for future research, especially as generative AI improves and becomes more accessible to a range of candidates.
Scaling deep learning for materials discovery
Amil Merchant et al.
Novel functional materials enable fundamental breakthroughs across technological applications from clean energy to information processing. From microchips to batteries and photovoltaics, discovery of inorganic crystals has been bottlenecked by expensive trial-and-error approaches. Concurrently, deep-learning models for language, vision and biology have showcased emergent predictive capabilities with increasing data and computation. Here we show that graph networks trained at scale can reach unprecedented levels of generalization, improving the efficiency of materials discovery by an order of magnitude. Building on 48,000 stable crystals identified in continuing studies, improved efficiency enables the discovery of 2.2 million structures below the current convex hull, many of which escaped previous human chemical intuition. Our work represents an order-of-magnitude expansion in stable materials known to humanity. Stable discoveries that are on the final convex hull will be made available to screen for technological applications, as we demonstrate for layered materials and solid-electrolyte candidates. Of the stable structures, 736 have already been independently experimentally realized. The scale and diversity of hundreds of millions of first-principles calculations also unlock modelling capabilities for downstream applications, leading in particular to highly accurate and robust learned interatomic potentials that can be used in condensed-phase molecular-dynamics simulations and high-fidelity zero-shot prediction of ionic conductivity.
Cynical, but useful? A lay beliefs perspective on cynical leaders' ability to prevent antisocial behavior at work
Teodora Spiridonova et al.
Social Psychology, October 2023, Pages 294-307
Cynicism -- the belief that people are driven primarily by self-interest -- has been predominantly associated with detrimental consequences for individuals and organizations. Less is known about its potentially positive implications. We investigated whether lay people consider cynicism helpful in preventing antisocial behavior and therefore see value in cynical leaders. We found that people viewed cynical (vs. trusting) leaders as better at detecting antisocial behavior and more punitive, and therefore, as better at preventing employees' antisocial behavior (Study 1). Despite this, cynical (vs. trusting) leaders were less likely to be hired, were offered lower salaries, and were seen as less effective (Study 2). This aversion to cynical leaders was attenuated for jobs that emphasized the importance of preventing antisocial behavior (Study 3).
Coming from a Good Pond: The Influence of a New Venture's Founding Ecosystem on Accelerator Performance
Administrative Science Quarterly, forthcoming
Startup accelerators, which aim to improve the set of choices representing a startup's entry strategy, have become increasingly influential in both regional development and the strategies of individual startups. This article explores an accelerator's impact on startup performance and whether that impact varies substantially by features of the startup's founding environment. Leveraging data from a leading startup accelerator, I use a regression discontinuity framework to hold startup quality constant so that I can compare the performance of admitted startups to those that do not make the cut, and I examine whether any observed performance differentials are driven by accelerator admission and by characteristics of the startup's earlier environment. I find evidence that startups from better pre-accelerator environments experience stronger gains from accelerator admission. I also find evidence of home bias, as local startups have a stronger treatment effect. These results provide evidence of ecosystem effects whereby the impact of one organizational sponsor in an ecosystem is strongly moderated by other features in the ecosystem. The findings help to explain the concentration of accelerator programs in already successful entrepreneurial ecosystems and reveal how such programs may interact with founding environments to complement resource abundance or magnify prior resource inequalities.