Digital technology, meritocracy and rent-seeking, experiment in China

How to cite

Chen, Junkai; Bu, Lingtian; Miao, Meng. 2025. "Digital technology, meritocracy and rent-seeking, experiment in China." AEA RCT Registry. AEARCTR-0015241. https://doi.org/10.1257/rct.15241-1.0

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

Junkai Chen* , Lingtian Bu , Meng Miao

* Primary investigator

Registry ID
AEARCTR-0015241
DOI
https://doi.org/10.1257/rct.15241-1.0
First registered
Last updated
Status
Completed
Trial dates
2024-01-01 – 2024-02-01
Countries
China ()
Registry record
http://www.socialscienceregistry.org/trials/15241

Abstract

The development of digital technology enables governments to collect vast amounts of data to enhance the quality of public services. Yet, concerns arise that such digital initiatives by governments might lead to increased power centralization and privacy infringement. In developing countries, such as China, local governments make significant investments in deploying digital technology, including the extensive installation of cameras for comprehensive street monitoring. This raises the question: What motivates government officials to actively support and implement these digital initiatives?

One explanation suggests leveraging digital technologies creates discreet rent-seeking opportunities. Top achievers in civil service examinations, ideally rewarded with higher pay according to traditionally held meritocratic principles, confront the stark reality of their modest salaries further diminishing due to government financial problems. This contrast, along with the potential efficiency gains from digital governance, makes rent-seeking justifiable.

We conduct a randomized experiment among 2,000 Chinese civil servants, applying two types of informational interventions—meritocracy and income. The meritocracy intervention inquires how many competitors they defeat in the civil service examination, whereas the income intervention asks if their recent salaries have been lowered. Civil servants then face a choice between fully automated and semi-automated devices for collecting data on enterprise fire safety. Compared with the fully automated device that autonomously uploads data to senior government levels, the semi-automated option requires manual intervention to decide on data uploads, creating rent-seeking opportunities.

Our findings indicate a preference for the semi-automated option under the income intervention, with a more pronounced pattern when combined with the meritocracy intervention. Our list experiment finds that these civil servants indeed understand that semi-automated systems offer more opportunities for rent-seeking, and they show greater tolerance for corruption after the intervention. We also dismiss alternative explanations such as technology idolization, distrust in technology, desires for power, and pro-social behaviors.

Experimental design

We conduct a randomized experiment among 2,000 Chinese civil servants, applying two types of informational interventions—meritocracy and income. The meritocracy intervention inquires how many competitors they defeat in the civil service examination, whereas the income intervention asks if their recent salaries have been lowered. Civil servants then face a choice between fully automated and semi-automated devices for collecting data on enterprise fire safety. Compared with the fully automated device that autonomously uploads data to senior government levels, the semi-automated option requires manual intervention to decide on data uploads, creating rent-seeking opportunities.

Design details

Randomization method
Randomization is done by a computer
Randomization unit
Respeondent
Clusters
2000 respondents
Observations
3000 respondents
Arms
1000 respondents in experimental group and 1000 respondents in conytol group
Primary outcomes
Our findings indicate a preference for the semi-automated option under the income intervention, with a more pronounced pattern when combined with the meritocracy intervention. Our list experiment finds that these civil servants indeed understand that semi-automated systems offer more opportunities for rent-seeking, and they show greater tolerance for corruption after the intervention. We also dismiss alternative explanations such as technology idolization, distrust in technology, desires for power, and pro-social behaviors.

Cited by

No citation data available for this registration yet.

Keywords