Digital Centralization and Autocracy: Evidence from Satellite Land Monitoring in China

Last registered on January 22, 2025

Pre-Trial

Trial Information

General Information

Title
Digital Centralization and Autocracy: Evidence from Satellite Land Monitoring in China
RCT ID
AEARCTR-0015242
Initial registration date
January 20, 2025

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
January 22, 2025, 8:27 AM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Region

Primary Investigator

Affiliation
Renmin University of China

Other Primary Investigator(s)

Additional Trial Information

Status
Completed
Start date
2024-07-01
End date
2024-08-01
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
The development of digital technologies represents a double-edged sword in the field of public governance. While rapid advancements in digital technology enable governments to access more real-time data, such data does not always accurately reflect reality. When government officials rely excessively on data to make decisions, they may overlook the complex contexts behind the numbers, leading to centralized and arbitrary decision-making, reduced administrative efficiency, and distorted social welfare.
This paper focuses on the negative impacts of digital technology on administrative systems in authoritarian states. We propose the hypothesis that big data can lead officials to make decisions with overconfidence in data, exacerbating centralization and authoritarian tendencies. When lower-level governments recognize the inefficiency and arbitrariness of such decision-making, they may resort to deception to correct these distortions and prevent misguided policies from being implemented.
Using a real-world example, we simulate the use of satellite digital information systems for land type monitoring. In China, local governments are tasked with ensuring that the proportion of arable land does not fall below the "cultivated land red line." Satellites can periodically capture images of designated areas to monitor land use types, enabling higher-level governments to supervise the performance of lower-level governments in fulfilling this mandate.
However, the data provided by such systems does not always reflect on-the-ground realities. For various reasons, some plots of land may become unsuitable for cultivation, prompting local officials to make flexible adjustments by designating equivalent plots of land elsewhere as substitutes. In these cases, if higher-level governments rigidly demand rectification based solely on satellite data, it can result in unnecessary resource waste.
We conduct a randomized experiment involving over 2,000 Chinese public officials, dividing participants into two groups: an upper-level government group and a lower-level government group. Participants in the upper-level group receive information to issue commands, while those in the lower-level group respond to the commands issued by their designated upper-level counterpart.
In the upper-level group, we randomly inform some participants about the existence of the satellite digital information system and show them satellite images of land plots, enabling them to use satellite-provided data for decision-making. Participants are given three options for issuing commands to lower-level officials, allowing us to observe whether they prefer lenient or strict approaches. We also measure the time they allocate for task completion and their responses to feedback from lower-level officials to capture multiple dimensions of their authoritarian tendencies.
In the lower-level group, some participants are randomly told that the upper-level officials are using the satellite digital information system for decision-making, while all participants are informed of the local challenges faced in land management. Lower-level participants are given three options: deceive the upper-level, report the real situation, or carry out the assigned task regardless of local realities. If they choose to report the truth, we simulate an upper-level rejection of their justification and observe how many rejections it takes before they stop reporting the truth.
Our findings reveal that access to digital technologies makes government officials more authoritarian and centralized in their decision-making. Participants with access to satellite data are more likely to issue strict commands. This authoritarian behavior stems from two key mechanisms: (1) the satellite project represents the authority granted by the state, and (2) the data itself leads participants to overestimate their informational advantage.
Among lower-level officials, knowing that upper-level decisions are based on satellite data initially prompts them to report the truth more diligently. However, upon realizing that their appeals cannot sway the upper-level, participants increasingly resort to deceptive tactics to evade tasks, rather than strictly implementing the upper-level commands.
We also compare the responses of public officials with ordinary citizens placed in similar scenarios. The results show no significant differences between the two groups when ordinary citizens are placed in the role of government officials. This indicates that the mechanisms we identify are not driven by characteristics unique to public officials but reflect a broader systemic issue inherent in the institutional environment.
External Link(s)

Registration Citation

Citation
Chen, Junkai. 2025. "Digital Centralization and Autocracy: Evidence from Satellite Land Monitoring in China." AEA RCT Registry. January 22. https://doi.org/10.1257/rct.15242-1.0
Experimental Details

Interventions

Intervention(s)
Intervention (Hidden)
Intervention Start Date
2024-07-10
Intervention End Date
2024-07-20

Primary Outcomes

Primary Outcomes (end points)
Our findings reveal that access to digital technologies makes government officials more authoritarian and centralized in their decision-making. Participants with access to satellite data are more likely to issue strict commands. This authoritarian behavior stems from two key mechanisms: (1) the satellite project represents the authority granted by the state, and (2) the data itself leads participants to overestimate their informational advantage.
Among lower-level officials, knowing that upper-level decisions are based on satellite data initially prompts them to report the truth more diligently. However, upon realizing that their appeals cannot sway the upper-level, participants increasingly resort to deceptive tactics to evade tasks, rather than strictly implementing the upper-level commands.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
We conduct a randomized experiment involving over 2,000 Chinese public officials, dividing participants into two groups: an upper-level government group and a lower-level government group. Participants in the upper-level group receive information to issue commands, while those in the lower-level group respond to the commands issued by their designated upper-level counterpart.
In the upper-level group, we randomly inform some participants about the existence of the satellite digital information system and show them satellite images of land plots, enabling them to use satellite-provided data for decision-making. Participants are given three options for issuing commands to lower-level officials, allowing us to observe whether they prefer lenient or strict approaches. We also measure the time they allocate for task completion and their responses to feedback from lower-level officials to capture multiple dimensions of their authoritarian tendencies.
In the lower-level group, some participants are randomly told that the upper-level officials are using the satellite digital information system for decision-making, while all participants are informed of the local challenges faced in land management. Lower-level participants are given three options: deceive the upper-level, report the real situation, or carry out the assigned task regardless of local realities. If they choose to report the truth, we simulate an upper-level rejection of their justification and observe how many rejections it takes before they stop reporting the truth.
Experimental Design Details
Randomization Method
Randomization is made by computer
Randomization Unit
REspondent
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
2000 respondents.
Sample size: planned number of observations
3000 respondents.
Sample size (or number of clusters) by treatment arms
1000 respondents in control group, 1000 respondents in experimental group and 1000 respondents in placebo test group.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials