Information Frictions, Algorithmic Matching, and Worker Preference: A Two-Stage Randomized Control Trial on Turnover and Productivity in Manufacturing

How to cite

Lin, Wei; Chen, Zixu; Chen, Ziyue. 2026. "Information Frictions, Algorithmic Matching, and Worker Preference: A Two-Stage Randomized Control Trial on Turnover and Productivity in Manufacturing." AEA RCT Registry. AEARCTR-0017986. https://doi.org/10.1257/rct.17986-1.4

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Wei Lin* , Zixu Chen , Ziyue Chen

* Primary investigator

Registry ID
AEARCTR-0017986
DOI
https://doi.org/10.1257/rct.17986-1.4
First registered
Last updated
Status
In development
Trial dates
2026-04-24 – 2029-04-30
Countries
Private
JEL codes
J63, M51, M54, O15, C93
Registry record
http://www.socialscienceregistry.org/trials/17986

Abstract

High employee turnover remains a persistent barrier to productivity in manufacturing firms. This study investigates the roles of pre-employment information, financial growth incentives, and job assignment mechanisms in mitigating turnover and improving worker-job matches. We conducted a large-scale field experiment with approximately 2,000 newly hired workers in a Chinese electronics factory using a 3 × 3 cross-randomized design. In the hiring stage, we test the effects of information frictions and perceived career growth on selection. Applicants are randomized into: (1) a pure control group; (2) an information treatment viewing a video on factory life and strict workplace regulations; or (3) a "growth path" treatment combining the video with information on on factory life, position-specific allowances, and tenure-based subsidies. In the job assignment stage, we analyze the efficiency of managerial decision-making versus data-driven and preference-based approaches. Workers are cross-randomized into three assignment protocols where managers receive different information sets: (1) random assignment; (2) improved matching-based assignment, utilizing an algorithm that predicts retention and productivity based on worker attributes; or (3) preference-based assignment, where managers are informed of workers' stated task preferences. We measure outcomes including retention rates, productivity, and job satisfaction to determine optimal strategies for personnel management in high-turnover environments.

Intervention

We collaborate with the in-house factory of a Chinese technology firm manufacturing earphones. The factory currently faces a critical challenge: high turnover rates among general workers and a disconnect between worker capabilities and task assignment. The production line operates under a hierarchy (General Workers, Multi-skilled, Leaders), but recruitment is rapid and non-selective. Managers currently assign tasks based on limited heuristics (e.g., arrival order, gender, perceived energy), leading to potential mismatches.

To address these frictions, we propose a two-stage field experiment involving a projected cohort of 2,600 new hires. The design disentangles supply-side mechanisms (worker expectations and incentives) from demand-side mechanisms (managerial matching quality).

Experimental design

This study employs a 3x3 cross-randomized design to disentangle the drivers of factory turnover and productivity through two distinct stages. In the Hiring Stage, recruits are randomized into a control group, a "Realistic Job Preview" group viewing a video on strict factory life to align expectations, or a "Task-specific Compensation Incentive" group receiving the video plus details on long-term wage incentives. Subsequently, in the Job Assignment Stage, managers assign these workers to tasks using one of three protocols: random assignment, an AI-assisted method utilizing algorithmic predictions of retention and productivity, or a Preference-based method incorporating workers' stated task desires. By interacting supply-side information treatments with demand-side matching mechanisms, the study tests whether retention is better improved by screening for fit and incentives or by optimizing the manager's assignment decisions through data and worker voice.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Individual
Clusters
No
Observations
2000 newly hired workers
Arms
2600 newly hired workers
Primary outcomes
1. Retention: Survival analysis of tenure (days worked). 2. Productivity: Output per hour and quality control (error rates) normalized by task type. 3. Attendance: Absenteeism and lateness. 4. Compliance: The rate at which managers follow the AI recommendation or the Worker Preference (checking for principal-agent friction between researchers and floor managers). 5. Subjective Well-being: Self-reported job satisfaction and alignment with expectations (collected via follow-up surveys).

Cited by

No citation data available for this registration yet.

Keywords