Dancing with anklets: Navigating digital rural governance with big data in China

Abstract

The digital village strategy is a critical component of China's national blueprint for integrating big data and digital technologies into rural revitalization. This study investigates the implementation of digital grid governance in rural China through a case study of County F—a national pilot site for the Digital Village Strategy. The study highlights the flexible, experimental, and iterative nature of local implementation, shaped by bureaucratic constraints and grassroots realities. The findings illustrate that digital governance entails significant financial and labor investments, particularly regarding frontline actors tasked with continual data collection and reporting. The study emphasizes the central role of relational labor in sustaining digital systems—an often invisible effort rooted in trust, familiarity, and embeddedness within local social networks. Such labor helps local officials navigate tensions between top-down mandates and rural complexities. Digital governance functions more as a symbolic performance of administrative achievement than as a tool for effective governance. The metaphor of “dancing with anklets” captures the politics of ambivalence under digital constraints.

This article is a part of special theme on Infrastructure, Labour, & Social Change. To see a full list of all articles in this special theme, please click here: https://journals.sagepub.com/page/bds/collections/infrastructure_labour_social_change?pbEditor=true

Introduction

China has increasingly integrated digital technologies, including intelligent sensors, big data platforms, and AI recognition systems, into various governance domains. Initiatives such as Smart Cities and Future Communities are significantly enhancing frontline governance in urban China, improving administrative efficiency and public service delivery (Große-Bley and Kostka, 2021; Zhao, 2023). Building on the demonstrated effectiveness of digital governance in urban communities and the growing penetration of digital technologies in rural areas, the state launched the Digital Village Strategy in 2019. The strategy has become an essential part of China's blueprint for integrating big data and digital technologies, driving economic, agricultural, and governance transformation in rural areas.

The effectiveness of digital governance depends on local resources, funding, skilled personnel, organizational capacity, and regional geographic and demographic factors, making it a highly localized rather than a unified nationwide system (Mittelstaedt, 2023). Studies have examined the digital transformation of frontline governance in Shenzhen, Zhejiang, and other urban middle-class neighborhoods (Tang, 2020; Große-Bley and Kostka, 2021; Xu and He, 2022; Zhao, 2023). A central instrument in this transformation is the grid system, which divides communities into small administrative units and collects detailed data about residents, housing, and land through local digital platforms (Mittelstaedt, 2022). In this way, digital grid systems (DGS) link individuals to their living spaces and automatically summarize information for reporting, monitoring, and control.

However, rural China faces a digital divide in infrastructure and digital literacy compared with urban areas. Moreover, the localized traditions of rural self-governance pose challenges to the implementation of standardized digital governance systems (Xu and He, 2022). Rural areas thus exhibit distinctive state-society relationships and sociocultural conditions (Liu, 2022) that shape unique digitalization processes. As Dencik et al. (2019) suggest, the heterogeneity of data practices demands a nuanced understanding of how data systems are contingent on specific contextual factors. Adopting this perspective requires moving beyond a purely technical engagement with digital technologies toward a broader political analysis of their development, deployment, and everyday implementation. This study investigates the implementation of China's digital rural governance in Pilot County F. It examines how local governments and frontline actors in rural China navigate digital governance mandates amid institutional and resource constraints, exploring how local communities engage with, negotiate, and reshape digital technologies in their everyday governance practices.

Literature review

Digitalization of frontline governance

Data-driven public administration is increasingly viewed as a means of efficiently targeting public services, understanding social problems, and allocating resources amid spending cuts (Dencik et al., 2019). It aims to predict and pre-empt activities and outcomes for more proactive citizen engagement (Dencik et al., 2019; Habich-Sobiegalla et al., 2024). Yet the ways in which data systems are developed, integrated, and enacted in practice—and their implications for frontline administrators—remain underexplored. China's ongoing digital transformation provides a telling example through which to examine these dynamics.

China's digital governance model is state-led; the government has actively adopted digital technologies to improve administrative efficiency and enhance oversight, thus facilitating better service delivery and greater transparency (Gao and Tan, 2020; Große-Bley and Kostka, 2021; Huang et al., 2024). In addition to improving public services for citizens’ daily interactions with the government, some cities, such as Beijing, have sought to address diverse and fragmented demands through the digitalization of frontline governance to maintain social stability (Habich-Sobiegalla et al., 2024). Frontline governance, the lowest level of administration in China, is often referred to as grassroots governance within the Chinese administrative system. In practice, grassroots governance primarily refers to administrative activities at the village and urban community levels. Although officially advocating co-governance and encouraging local societal participation, this type of governance remains largely state-led in practice (Mittelstaedt, 2022; Tang, 2020; Xu and He, 2022).

Maintaining social stability has long been a central task of the Chinese state (O'Brien and Li, 1999; Edin, 2003; Chen and Liu, 2021). Beyond passively responding to citizens’ requests (e.g., through citizen hotlines and the petition system), the state has continually expanded its infrastructural power through technological innovation and institutional development (Chen and Liu, 2021). At the grassroots level, grid governance complements existing information collection channels by proactively managing residents within grids while preventing or resolving neighborhood conflicts (Xu and He, 2022). Unlike citizen-initiated appeals typically recorded at the municipal level, grid governance relies on grid workers to actively collect data; this data is stored in county-level command centers (Habich-Sobiegalla and Plümmer, 2023). Through this process, the state seeks to establish a data-driven “proactive governance” system (Habich-Sobiegalla et al., 2024). DGS exemplifies the datafication of frontline governance. It involves transforming various aspects of grid-based life, such as people, housing, and events, into quantifiable data for tracking, analysis, and decision-making.

Framed as a “grassroots governance innovation,” DGS has been widely implemented in urban China. Tang (2020) suggests that the grid governance system has become a crucial mechanism for mobilizing residents, resolving conflicts, and reinforcing Party leadership in these neighborhoods by integrating resident volunteers, utilizing the political influence of retired urban elites, and establishing Party organizations within residential communities. While it enhances the accessibility, efficiency, and fairness of public services for citizens, Mittelstaedt (2022) considers that a top-down administrative approach that disregards people's agency may weaken local engagement and hinder collaboration with other stakeholders. Lei (2023) employs the metaphor of a “gilded cage” to illustrate the complex, paradoxical nature of China's technological development and state-led governance transformation; this highlights the tension between enhanced digital governance capabilities and the constraints imposed by centralized control. However, while these dynamics have been extensively studied in urban areas, the development, implementation, and effects of DGS in rural China—where resources, infrastructure, and local governance traditions differ—remain underexplored.

The context of rural China

The effectiveness of digital governance is determined not by objective technology but by “enacted technology” (Gao and Tan, 2020). Unlike urban communities, characterized by heterogeneous and transient social structures (Mittelstaedt, 2023; Tang, 2020), villages comprise emotional and life-sharing communities, with social structures built upon kinship and geographical relationships (Chen and Liu, 2021; Wong et al., 2020). The intimate nature of production and daily life in rural areas has led to the formation of strong informal norms, which serve as a solid foundation for rural autonomy. Informal institutions and traditions indigenous to rural communities (e.g., collective identity, shared norms regarding kinship relations, and perceptions of authority and legitimacy) continue to influence daily interactions among residents (Chen and Liu, 2021; Wong et al., 2020). Even in urbanized villages or villages in cities with high concentrations of migrant workers, grassroots governments are often reluctant to intervene in village affairs (Xu and He, 2022). Therefore, as Liu (2022) argues, research must extend beyond an organizational-centric perspective and consider the broader political and cultural contexts in which data and governance interact.

Drawing on the Social Data Governance Model (Liu, 2022), urban and rural China represent markedly different contexts. Regarding “state–society relations,” rural areas are shaped by traditions of local autonomy and embedded social ties (Xu and He, 2022). In comparison, heterogeneous and mobile urban populations tend to rely more on digital technologies for direct state interactions (Tang, 2020). Based on cost–benefit calculus, Kostka (2023) indicates that the adoption of digital technologies in China is generally supported by wealthier, higher-educated, and urban residents. Given the considerable barriers to direct interaction between the state and villagers, village cadres often enjoy a near-monopoly on upward and downward information flow, granting them significant discretion in managing various everyday issues (Chen and Liu, 2021). Culturally, rural society is rooted in collectivism and dense kinship networks, in contrast to the more individualistic orientation of urban settings. Rural grid workers are typically village cadres or collective staff. Their embeddedness in local networks is key to building trust and resolving conflicts.

Amid the growing prevalence of digital technology in Chinese villages, the state has increasingly promoted the use of digital technologies to enhance public services and oversight of frontline officials, aiming to mitigate potential conflicts arising from rural autonomy (Xu and He, 2022). Grid workers are therefore caught in a “digitalized iron cage” with pre-programmed roles and rules (Gao and Tan, 2020). However, implementing these digital initiatives relies heavily on grid workers to collect information and respond to local requests (Zhao, 2023; Habich-Sobiegalla et al., 2024). Grid workers depend on their informal social networks and situational adaptations to upload localized or “indigenized” data that existing sensor technologies are unable to capture (Zhao, 2023). Addressing residents’ requests not only involves procedural responsibilities for grid workers but often demands emotional labor to navigate tension and maintain trust (Zhang, 2025). Consequently, these frontline officials are both controlled subjects and active agents shaping control mechanisms.

Dancing with anklets

Both the “iron cage” (Gao and Tan, 2020; Zhao, 2023) and “gilded cage” (Lei, 2023) seemingly presuppose the effective implementation of digital governance. Under the cage metaphor, frontline officials are viewed primarily as objects of surveillance that is intended to improve public service delivery. Digital governance may reinforce ethical and democratic values by making decision-making more transparent, consistent, and accountable (Busch and Henriksen, 2018). Furthermore, data-driven governance often features characteristics such as standardization, proceduralization, transparency, and objective verifiability (Xu and He, 2022). However, it leads to a form of technological solutionism that overlooks context, people, values, and history, treating the local community as a technical system rather than a multifaceted entity (Zhao, 2023). Thus, digital governance may undermine local knowledge and relational values by limiting frontline officials’ capacity to use personal judgment, expertise, and direct interactions with residents (Busch and Henriksen, 2018).

However, village officials in China are not merely policy implementers. They are bureaucrat–politicians tasked with the full spectrum of rural governance (O'Brien and Li, 1999). Their policy implementation is often selective (O'Brien and Li, 1999) or prioritized (Edin, 2003), demonstrating that they are not entirely constrained by rigid bureaucratic structures. Moreover, the effectiveness of policy implementation largely relies on informal mechanisms, including personal networks, kinship ties, trust, and verbal negotiation (Zhao, 2023). To capture the tension between top-down administrative control and local strategic flexibility, this study employs the metaphor of “dancing with anklets.” Here, “anklets” symbolize hierarchical administrative constraints and digital surveillance within digital governance, and “the dance” represents the agency of local governments and village officials as they navigate and negotiate such constraints. This analytical framework unfolds along three dimensions: performative compliance, constrained improvisation and relational labor dependence.

First, dance is inherently performative. Promoting digital governance enables local governments to perform local innovation and exemplary status (Zhao, 2023). Implementing digital rural governance is not a standalone central policy; it is closely tied to broader national agendas such as Digital China, the Digital Village, Rural Revitalization, and the modernization of governance capacity. It can be viewed as a “priority policy” (Edin, 2003). However, its objectives and pathways remain uncertain and loosely defined. For such innovative policies, policy pilots serve as mechanisms for formulation, experimentation, and iterative adjustment to clarify goals and reduce uncertainty (Zeigermann and Ettelt, 2023). Furthermore, the central government encourages localities to explore context-specific approaches and generate “experiences” that can serve as references for digital village development elsewhere. Against this backdrop, being selected as a digital village pilot allows local governments to performatively signal their technological agility and political responsiveness (Zhao, 2023). Additionally, under the cadre rotation system, local leaders are incentivized to pursue quick, visible results in order to become model cases, which can facilitate career advancement and performance-based rewards (Huang, 2002; Edin, 2003).

Second, the metaphor of dance highlights not only structured coordination but also interaction and improvisation—the unexpected elements that arise during performance. The ambiguity of policy goals and pathways, along with a lack of clear precedents, grants county governments considerable autonomy (Chen et al., 2024). Uncertainty within the policy piloting process compels local governments to interpret and perform the “choreography” of central directives in ways that reflect their specific implementation contexts (Howlett, 2018). As Mittelstaedt (2023) notes, the effectiveness of grid governance relies heavily on the availability of local resources and expertise. Thus, policy implementation becomes a situated, improvisational practice, shaped by local constraints, interpretations, and adaptive strategies.

Finally, the uncertainty of localized contexts means that the implementation of digital rural governance also depends on the “dancers”—frontline actors—and how they enact their roles and responsibilities within everyday interactions. For instance, Zhang (2025) examined the experience of female street-level bureaucrats’ emotional labor in defusing conflicts, nurturing short-term stability, and projecting a caring image of the state. In contrast to the more incidental interactions between urban frontline officials and citizens, rural officials maintain more enduring and routinized relationships with villagers. Villagers are not merely perceived as “users” and reduced to data points within digital systems (Broomfield and Reutter, 2022); they are also acquaintances, friends, neighbors, and even relatives.

Rather than merely managing feelings during isolated encounters, frontline actors engage in what Baym (2015) terms relational labor—ongoing, effortful communication over time that sustains trust, mutual obligation, and cooperation in support of institutional goals. Unlike emotional labor that serves to manage others’ feelings during single encounters (Zhang, 2025), relational labor builds over time through ongoing connections (Baym, 2015). As “mobile eyes” (Xu and He, 2022), frontline actors are embedded in local social networks, leveraging emotional and relational ties to mediate daily conflicts and maintain reciprocal trust with both higher authorities and villagers. Thus, relational labor not only enhances credibility and effectiveness but also provides the affective infrastructure upon which digital governance quietly depends (Chen and Liu, 2021; Wong et al., 2020).

Following the methodology section, this study examines how China's Digital Village Strategy has been selectively implemented in County F. It analyzes how local authorities adapt central directives to local conditions, highlighting the flexibility and iterative adjustments involved as experimental policies move from broad mandates to concrete action. Second, it investigates how the DGS—metaphorically described as “digital anklets”—is built in practice. The analysis traces the process of data collection, integration, analysis, and decision-making, showing how local governments manage to assemble functional digital infrastructure despite limited resources and institutional constraints. Third, the study explores the agency of frontline actors. It illustrates how performance pressure shifts the focus from effectiveness to visibility, and how digital systems become dependent on informal relationships for functionality. Finally, this paper offers a critical reflection on digital governance through the lens of this rural case study. While digital technologies are often seen as efficient and cost-saving, the Chinese experience reveals that such systems remain highly labor- and resource-intensive. Furthermore, the paternalistic nature of state-led investment not only reinforces existing social inequalities but also generates new forms of exclusion and disparity.

Methods

This study examines a case—County F—located in Shandong Province in eastern China, based on two rounds of fieldwork and thirty-five semi-structured interviews. In October 2020, the Cyberspace Administration of China, in collaboration with relevant ministries, issued a notice identifying the first group of national digital village pilot areas; the pilot aimed to summarize and promote effective digitalization practices, with County F selected as one of the first pilot sites. As a national pilot, the digital governance practice in County F offers a representative glimpse of the overall bureaucratic transformation, leading to empirical insights into specific characteristics of digital rural governance.

To implement the Digital Village Strategy, County F selected Township T and WB Village as model cases of digital rural governance. In August 2022, I visited these two model sites and interviewed nine informants, including two county leaders from the Cyberspace Administration of County F, one county official, two township officials, two village cadres, and two heads of technology companies involved in the region's digital village construction. The fieldwork focused mainly on how the county and township governments strategically implement national policy. Digital village practices are still in their early stages, having only established a digital platform framework and initiated the deployment of “smart sensor” equipment; integration and application of data collection and consolidation are in progress. Furthermore, I gradually understood that the Organization Department of County F is extending digital governance to rural areas through DGS, thereby improving frontline governance capacity. Specifically, each grid in County F is designed to cover an area with approximately 300 to 1000 permanent residents. An administrative village typically comprises three to four grids, each staffed by designated grid workers. The village Party Secretary serves as the grid leader, while other village officials act as grid workers, responsible for executing daily tasks.

In December 2024, I conducted additional fieldwork to explore the implementation practices of county and township governments, the daily digital practices of grid workers, and villagers’ perceptions of digital rural governance. I visited the county-level Digital Rural Big Data Platform and Grassroots Governance Command Platform while investigating 11 villages in six townships. I first went to the towns and villages where I conducted my first survey. Yet interviews with grid workers in different townships revealed inconsistencies in how “incident reports” were submitted, highlighting interpretative differences in implementing county-level guidelines. In light of these findings, I expanded the second round of fieldwork to include a broader range of townships and villages. This allowed me to better capture the diversity, fragmentation, and evolving dynamics of digital governance practices in County F. I observed how the county-level platform operates and accompanied grid workers during their patrols in villages. Moreover, I conducted twenty-six semi-structured interviews: three county leaders, two county officials, two township leaders, three township officials, one head of the county's big data platform center, twelve village cadres (four of whom are village party secretaries), and ten grid workers (five of whom are also village cadres). The interviews focused on how participants collect, integrate, and apply data in their daily work. Additionally, I interviewed eleven villagers, but only one was familiar with grid workers. Interviews with villagers concerned their awareness of digital governance, opinions on village cadres’ performance, and their views on local conflicts and disputes.

Except for Township S, where the higher-level government facilitated my access, I conducted my fieldwork independently as an academic. Frontline officials and grid workers were more open and candid without “political” pressure. In Township S, county and township officials spearheaded the visit and often “helped” village cadres and grid workers answer questions during the interviews. I collected their contact details and conducted follow-up interviews via WeChat. I disclosed the study's purpose and confidentiality at the start of each interview, assured participants of academic use only, and obtained informed consent. Participants requested anonymization of their locations and identities, so all place and individual names are pseudonyms; specific data is anonymized to prevent identification while maintaining authenticity.

All field observation notes, research diaries, and interview dialogues were recorded by the author and later reviewed and supplemented by two research assistants. The research data were inductively coded to identify and analyze pertinent themes related to the research questions. The county and township governments shared policy documents and local experience reports that do not contain confidential information. These secondary data sources provide valuable insights into imagination and knowledge production regarding the relationship between technology and governance.

Digital village implementation in County F

The Digital Village Strategy shares the same goals as smart cities (Große-Bley and Kostka, 2021) and future communities (Zhao, 2023), illustrating a state-led drive for data centralization and the creation of a unified digital platform infrastructure (Große-Bley and Kostka, 2021). This aims to enhance oversight of frontline governance and provide better public services (Xu and He, 2022). However, there is no existing experience or unified standard for implementation pathways. Characterized by a “crossing the river by feeling the stones” approach, the policy pilot has played an essential role in China's state governance. By constructing these various demonstrations, policy pilots aim to transform the “uncertainty” of contextual factors into the “certainty” of policy pathways (Chen et al., 2024).

China's Digital Village Strategy involves multiple government departments. The pilot project needs a dedicated team to carry out this work. With such initiatives, it is common for a Deputy Party Secretary of the county to serve as the team lead, responsible for coordinating interdepartmental collaboration. One specific department is designated to take the lead in organizing and executing the project: the Cyberspace Administration of County F (CAF), which coordinated overall implementation. At the same time, other departments—such as the Agriculture and Rural Affairs Bureau and the Bureau of Industry and Information Technology—are required to cooperate and participate in the process. Regarding the implementation plan, County F outlines the following tasks:

The national digital village pilot project has outlined ‘seven key tasks’, requiring regions to tailor their approaches to local conditions and prioritize areas that can yield quick results. In accordance with this directive, we have focused our innovation efforts on industrial digitization and digital rural governance. The goal is to develop an ‘F Model’ for digital rural development that is replicable, scalable, and can serve as a reference for other regions (F20220224).

Emphasis on tailoring to local conditions reflects significant contextual factors (Dencik et al., 2019; Mittelstaedt, 2023). However, requirements and goals, such as quick results and the creation of a replicable and scalable model, also highlight the constraining context. To reduce the ambiguity in policy objectives, policy pilots should allow for sufficient patience to enable redesign and recalibration (Chen et al., 2024). Yet the pilot project has a clear time limit. In County F, “quick results” means that a satisfactory outcome must be delivered within three years. This limited time frame poses challenges in mobilizing human and material resources. Specifically, County F faces significant funding challenges, with “substantial investment needed for digital village infrastructure, and relying solely on county-level fiscal input is a heavy burden” (F20230116).

The CAF has actively sought collaboration with various stakeholders. Given the limited project funding, several state-owned enterprises were mobilized to invest more than 20 million RMB to create a Digital Rural Big Data Platform (BDP). Since the platform was established, data collection and integration have become a significant challenge. For instance, data related to agriculture, rural areas, and rural populations remain scattered across more than a dozen relevant departments and companies. Initiating inter-organizational collaboration for data sharing across government departments faces significant obstacles—not only institutional, but also legal and technical (Gao and Tan, 2020; Große-Bley and Kostka, 2021). Therefore, while the big data platform was completed by the end of 2022, data collection and integration remain a work in progress.

Given the difficulty in quickly and effectively integrating horizontal departmental data to fully utilize the big data platform, the CAF shifted responsibility to township governments. Driven by higher-level authorities, the implementation of digital projects may be enforced through state hegemony or operate “in the shadow of hierarchy” (Mu et al., 2019). Furthermore, township governments are eager to be recognized as model cases to receive top evaluations during annual performance assessments. Achieving such recognition not only grants them higher performance-based bonuses—often amounting to two to three times their regular monthly salary—but also improves their chances for promotion (Edin, 2003). Specifically, Townships T and B separately implemented the Digital Town and Palm Village initiatives in WB village to establish model cases; however, these projects were entirely separate from the BDP. Township T and WB village are responsible for providing basic data. Digital system maintenance and data storage are managed by technology companies. Although they claim to promote data-driven governance (GY, Township leader), in practice, their implementation of big data has been limited to showcasing the installation of smart sensors, such as facial recognition locks and digital gate systems for identifying and registering residents and vehicles.

The digital platform and installation of these sensors provide a visual display of digital capabilities, showcasing their level of “intelligence.” This is one of the easiest achievements to implement within a short time frame. It also serves as a demonstration for higher-level authorities (Zhao, 2023). Additionally, smart sensors can generate profit for collaborating technology companies, as noted: “The cost of the sensors themselves, along with the subsequent maintenance fees, enables technology companies to generate sustainable revenue” (ZR, CAF leader). Through collaboration between frontline authorities and technology companies, model cases are created. However, model cases often serve as “digital achievements” for government and “digital products” for enterprises, transforming digital village initiatives into little more than a form of “digital adornment” for local governments.

With the completion of the digital village pilot project by the end of 2022, the digital governance practice in County F has stagnated. During the subsequent investigation, it was noted that “Township T's contract with Company M had expired, and the data had not been updated” (WL, CAF official). The smart sensors invested by Company M have not produced sustainable returns: “These are all rural communities in which villagers are unwilling to pay, so the company could not collect the fees, and now Company M is increasingly struggling” (ZR, CAF leader). As Mu et al. (2019) note, vertical governance typically occurs at the initial stage, with a higher authority pulling in necessary agencies to heighten perceived interdependence. At this stage, collaboration between local governments and enterprises could help to achieve their respective short-term goals. However, partnerships that overlook local contexts and rely solely on top-down authority are difficult to sustain. During my second field visit, GY had been promoted for outstanding performance in leading digital implementation in Township T. Unsustainable investment—though challenging to quantify in terms of its impact on local governments—can deal a fatal blow to businesses, particularly small local enterprises.

Completing the digital village pilot project does not signify the end of digitization in frontline governance. During the COVID-19 pandemic, the county government realized it lacked detailed and accurate population data at the community level. In the urgent context of epidemic prevention and maintaining social order, the previously performative and formalistic “digital decoration” had to be transformed into practical functionality. In response, County F integrated its existing grid governance system with digital technologies to promote digital grid governance.

Digital grid governance serves not only as a mechanism for delivering public services but also as a means of maintaining social stability. Therefore, County F tasked the Organization Department with taking the lead. As Lian (2018) argues, the authority status of the principal is a powerful incentive signal. In this context, the Organization Department, given its role in personnel promotions, holds a relatively high position and exerts informal influence over different departments. The Grassroots Governance Promotion Center (GGPC), under the Organization Department, implemented the DGS as a tool for frontline governance in both urban and rural areas. Leveraging its authority, GGPC can mobilize collaboration horizontally and vertically across departments, integrate data, and safeguard data security, thereby turning digital adornment into practical functionality. The following section explores how DGS was established and operationalized.

Wearing the anklets: Assembling the digital grid system

In the process of promoting digital grid governance, multi-layer digital applications of F shanzhi (善治, good governance) have been enhanced. GGPC leverages the command screen to coordinate interdepartmental collaboration in managing complex events, while concurrently providing data to support government decision-making. Township governments use web platforms to collect data and address issues that cannot be resolved at the community level. Grid workers record daily activities and report events via their “Pad.” Residents can seek assistance or submit complaints and suggestions via the app. The advancement of DGS illustrates the process of datafication in digital governance.

Data identification and gathering

Data lies at the core of digital governance; the integration of basic rural data and information resources serves as the foundation for data-driven governance (Große-Bley and Kostka, 2021). The system first requires collecting vast amounts of data on rural elements, establishing a foundational database organized around “people, sentiments, locations, objects, events, and organizations.” The database unifies existing information into special, easily manageable dossiers that both the grid and the authorities can access.

The GGPC integrated these data through two primary methods: data collection by grid workers, and interdepartmental data “interflow” (互通, hutong). Interflow refers to the flow of data occurring at a physical level, relying on human effort rather than the direct integration of data into the platform for cross-departmental data sharing.

These data are not directly integrated into the platform. For example, the Disability Federation provides a list of disabled individuals, which is periodically sent to the GGPC. The GGPC confirms and verifies this information with the township authorities, after which it updates the platform data. Grid workers’ data was also updated accordingly (LY, Township official).

Although the mobilization of the Organization Department has facilitated necessary cooperation among departments, fully adjusting the existing governance structure to align with digitization logic and integrate data presents significant challenges. Therefore, foundational data is primarily collected manually by frontline officials, not through digital systems.

These data can be classified into two types: static and dynamic. Static data primarily involves assigning digital identity codes to elements within the grid (e.g., population, housing, land, and rivers), thereby providing a panoramic digital representation of the overall rural landscape. Like urban grids, the importance of grid workers in collecting data has been emphasized by local governments (Zhao, 2023). Regular training sessions for grid workers were held at both the county and township levels. This training covers a wide range of topics, from role expectations and data security awareness to the practical use of grid management tablets, including step-by-step guidance on operational procedures (XM, County official).

Dynamic data primarily includes changes in basic information within the grid, as well as event information gathered during routine inspections. As the codification of people and structures was completed, the focus gradually shifted from data collection to data verification. To ensure the accuracy of foundational data, the GGPC has organized six rounds of “dazoufang” (大走访, comprehensive visits), serving as a form of campaign-style governance. As an irregular task beyond routine work, this uncertain approach places political pressure on frontline officials, thereby ensuring that all data within the grid is thoroughly inspected and verified for accuracy.

Moreover, conducting daily grid inspections and gathering intelligence are the primary tasks of grid workers. Grid workers are required to carry a Pad equipped with the “F shanzhi” app, which is designed for intelligence reporting. The app facilitates the collection of dynamic data and enables the command center to track workers’ inspection routes: namely, to ensure routine grid patrols and regular visits to the 27 categories of key individuals—such as people with disabilities, children in distress, and elderly people living alone. Collecting and reporting event information is another key duty of grid workers. According to county regulations, event information focuses on areas such as public sentiment, conflicts and disputes, fire safety, and environmental sanitation (F20230711).

Interviews with grid workers reveal very few conflicts and contradictions among the aging rural communities. Yet, at the initial stage of DGS implementation, the GGPC mandated the daily reporting of event data, even when no actual events occurred. Grid workers must fabricate incidents first and then purport to solve them: “I typically report issues such as fallen leaves on the roads or garbage that hasn’t been cleaned up in time” (CH, MJ village grid worker).

Inaccurate information can compromise the quality of data collection, thereby undermining the effectiveness of data governance (Mittelstaedt, 2023). To mitigate this, County F employs two methods: system-based analysis and supervision. The system issues warnings and deducts points through a false judgment system. However, the interviews reveal that false judgment essentially involves assessing whether the attribute category of an “event” is correctly classified (LY, Township official). However, determining whether an event has occurred remains challenging. As a result, township officials rely on their own experience to verify and report information. Reporting false or inaccurate data will affect the county's assessment of townships. Therefore, township officials not only supervise the data information of grid workers; they sometimes directly enter the data on their behalf to ensure the quality of the data, especially when entering data information based on maps and locations such as land and village boundaries, as described:

I drew village boundaries manually; due to the low level of education of the grid workers, I was unsure of the specific boundaries of these villages. They assisted me and provided me with specific information, and I was responsible for drawing the village boundaries. If grid workers drew them, the boundary line would be rough: the map boundary information we provided would not overlap with the boundaries of other township grids. It would be easy to see, and we would lose points in the assessment (LY, Township official).

Both grid workers and township officials invested considerable effort in comprehensively collecting foundational data for the entire village. While the BDP has already mapped land and village data, there has been no considerable integration of data across departmental and hierarchical levels. As a result, government agencies continue to rely on top-down mandates for repeated data collection, often duplicating efforts.

Data response mechanisms

In addition to routine maintenance, data retention, and analysis, events related to appeals, help-seeking, and consultation are also included in the various types of data collected on the platform. Event data enables local governments to promptly identify and respond to villagers’ demands while monitoring grassroots dynamics, thereby contributing to social stability. Reported issues are classified into three types based on their complexity: self-resolved cases, referred township cases, and referred county cases.

First, self-resolved cases can be handled directly by grid workers or village cadres. Grid workers indicated that almost all events were self-resolved, with villagers rarely initiating reports themselves. Grid workers often respond to villagers’ needs during daily inspections, addressing minor physical issues such as burst water pipes or fallen electrical cables. Issues that cannot be resolved independently are reported to the village committee for resolution. Second, township-level governance center cases involve events that require escalation to the township level for resolution through inter-departmental coordination. Facing complex issues that villages cannot resolve, grid workers are generally allowed to upload cases after receiving permission from the village cadres. The township grid center assigns events to the relevant parties responsible according to departmental responsibilities. During field research, such complex events were often related to struggling for resources from higher authorities, such as those who want to be identified as “five-guaranteed households” ¹ or disabled people (FJ, Township official). Third, in cases referred to the GGPC, if the township grid center is still unable to solve the problem, it will be escalated to the county center for resolution through departmental coordination. During field research, township and village grid workers did not report any complex events that required the county center to coordinate with various departments.

The DGS enables county and township governments to effectively access data about basic rural elements such as population, land, and housing, while monitoring and regulating frontline officials to provide more proactive services to villagers. The “anklets” metaphor illustrates how higher-level authorities can more precisely access data from grassroots society, strengthening oversight of local cadres in their efforts to better serve villagers. This supervision not only improves service attitudes but—more importantly—encourages proactive engagement with communities and the prevention of conflicts, thereby maintaining social stability at the grassroots level (Mittelstaedt, 2022; Xu and He, 2022). The shift from a management-oriented to a service-oriented approach among village officials reflects the administrative constraints embedded in a top-down digital governance system.

Strategic data application

Even the most accurate information is worthless without the ability to centralize, integrate, and distribute it (Große-Bley and Kostka, 2021). In County F, once data is collected, it is centralized in grid management centers, where it is analyzed and distributed. For all types of basic data and event information gathered, the digital governance centers at the county and township levels conduct regular analysis and judgment to serve government decision-making. The GGPC command screen terminal features an intelligent analysis and judgment function, which can summarize and identify common problems that must be addressed, especially for cases with more individualized demands. As GGPC leader JZ explained, “We identified 22 specific issues across six domains by analyzing 91 subcategories of commonly raised problems grouped into 10 major categories. Based on this analysis, we submitted a policy report to the county party committee.”

By visualizing and categorizing data trends, the DGS provides a more intuitive basis for governance. For instance, Township official FJ noted that “different colors on the platform's pie charts indicate which issues are most frequent, such as problems with the living environment. Based on nearly 200 such cases analyzed over the year, we launched the Humanity Settlement Environmental Improvement Actions; the number of related complaints has since dropped to fewer than a dozen.” Thus, the system enables not only responsive but also proactive governance, identifying issues before they escalate. However, this data-driven model has intensified the burden on frontline officials. As Habich-Sobiegalla et al. (2024) indicates, local cadres must now handle routine administrative work in addition to tasks triggered by system-generated analyses. The logic of performance through data turns frontline officials into executors of algorithmically assigned responsibilities, expanding their workload and increasing pressure to meet targets derived from abstract indicators.

Moreover, the system utilizes data integration and analysis technologies to collate data and automatically generate analytical reports. However, the reports often serve only as a reference or justification for decision-making, rather than being the core driving force behind it. For example, GGPC official XM noted, “Using data regarding the elderly population and restaurant distribution, we created a 15-min living service circle. Based on the report and the data on vacant houses from the platform, SZ village established a community canteen and a day care center for the elderly.” However, fieldwork revealed that the village collective already had sufficient income and had long planned to establish such facilities. Thus, the digital system was used instrumentally—to legitimize existing decisions rather than to initiate or shape them. The data report generated by the system merely provided a visual rationale for actions already underway. In many cases, local actors strategically utilize the system to align their priorities with higher-level expectations.

Performing the dance: Institutional contradictions and relational labor

DGS provides an essential form of infrastructural power for collecting residents’ basic information while addressing community affairs: “resolving small disputes within the grid and major conflicts within the community.” However, its underlying goal is to reduce the number of conflicts that are reported across administrative levels, e.g., bypassing village authorities and directly reporting to the city level through the 12345-citizen hotline. GGPC leader JZ said that assessment standards are no longer based solely on the number of reported events, but on “two decreases and one rise.” Two decreases refer to the reduction of citizen complaint events via the hotline and petition events, which imply that frontline incidents are not escalated to the county or city level. One rise refers to the number of events reported by grid workers, meaning that grid workers are encouraged to report issues that can be resolved internally. The assessment requirements illustrate that the DGS serves as a tool for county and township governments to minimize accountability pressure from the city level. Habich-Sobiegalla et al. (2024) study on Beijing's Jisujiban system illustrates how such cross-level reporting imposes administrative pressure on frontline governance. When residents bypass lower-level authorities, it exposes governance failures at the frontline level, triggering performance assessments by the county toward township governments.

Therefore, beyond supervision and service delivery, the fundamental purpose of DGS lies in preempting and resolving social conflicts to ensure social stability. Although the GGPC has canceled event quantity regulation, townships still maintain some “quantity” requirements for grid workers to obtain higher scores in the “two decrease one rise” assessment. Some townships stipulate that grid workers must report 12 events of conflicts and disputes each month, while some stipulate four to five events. Grid workers in Township T are required to report one routine work record daily, as well as 12 events related to conflicts and disputes each month. When asked how many conflict and dispute events were “created,” a grid worker responded that “Almost all of them. There are not many conflicts in the village, and some of the significant conflicts cannot be resolved” (DJ, ZL village grid worker).

Additionally, the case resolution rate is a crucial metric for assessment. The “rate” serves as a key indicator for local governments to evaluate the effectiveness of grid governance. As one township leader WQ noted: ‘The key is not to identify problems, but to find problems that can be solved. If cases such as “demands for unpaid salary” are reported and we cannot resolve them, they should not be reported.’ Rights protection incidents are highly politically sensitive, and the issue of wage arrears involves enterprises and the finance sector, which is difficult for township governments to address. Beyond deliberate withholding of sensitive or problematic events, grid workers also downplay the significance of conflict-related issues by manipulating the types of issues, reporting events that are politically uncontroversial and easier to resolve. Although various types of issues are reported, about 90% concern the living environment, followed by social assistance and basic medical care, representing less sensitive items (LY, Township official). Therefore, to complete the rate assessment, grid workers often report insignificant events, such as “picking up leaves” or “cleaning up garbage.” Thus, the metaphor of the “dance” reflects how local actors demonstrate their agency through ambiguous, diverse, and flexible responses (Tang, 2020).

Based on the above, frontline authorities and grid workers, through selective and repetitive creation of event reports, manage to “dance” with the anklets. In this interactive process, township governments both impose increasingly demanding tasks with higher quantitative expectations on village committees and grid workers and collaborate with villages to selectively fabricate repetitive event data. Village actors are compelled to report data as required by higher authorities to better serve the public, and respond to these demands by creating insignificant or easily resolvable events. Therefore, the digital anklet does not, as some studies suggest, diminish the agency of the people (Mittelstaedt, 2023). Instead, the DGS fosters a politics of ambivalence between oversight and agency, where both higher-level authorities and local actors may have competing goals and interpretations of what governance should look like. The politics of ambivalence in digital rural governance may not directly challenge established norms but destabilize them, generating an alternative space of intelligibility.

Such ambivalent politics ultimately render the realization of digital grid governance heavily dependent on frontline actors. It is their everyday labor—both technical and relational—that sustains the system's functionality amid institutional tensions, fragmented mandates, and resource constraints. Rural grid workers are often Public Welfare Post (PWP) ² workers or village cadres who serve concurrent roles. Village cadres have numerous responsibilities, and the specific tasks within the grid are usually carried out by PWP. These PWP-grid workers are hired and subordinate to the village committee, and the government pays them 800 RMB per month. In practice, when grid workers encounter issues that cannot be resolved on-site, they initially report these events to village cadres.

Relational labor is essential in the interactions between township officials and village cadres-grid workers. Their relationship is not simply hierarchical; in the face of mandates from county-level authorities, they engage in ongoing negotiation and coordination, built upon long-term interpersonal familiarity and mutual trust. As part of efforts to train grid workers and improve their operational effectiveness, GGPC has introduced initiatives such as “training-through-competition.” Furthermore, GGPC competitions are often held for “excellent grid workers,” in the form of individual speeches explaining their daily work. Village cadres—usually younger and better educated than PWP workers—are better able to win honors for the township in these competitions. As a result, township officials rely on personal relationships to cultivate young village cadres, molding them into “outstanding grid workers.” As Chen and Liu (2021) note, this type of relationship transforms formal government tasks into personal obligations. The relational connection between township and village interaction can weaken or even dissolve the top-down administrative control of the digital grid. Building on administrative employment and emotional ties, township and village actors dynamically manipulate the digital anklet through personal obligations.

Moreover, relational labor is evident in the everyday interactions between frontline actors and villagers. Through long-term interaction and daily encounters, trust is gradually built and facilitates policy implementation and conflict resolution. Sustained interpersonal relationships enable grid workers to informally and efficiently respond to villagers’ needs during routine inspections. This relational maintenance also serves to regulate villagers’ behavior. As one grid worker, who also serves as a village cadre, TT noted: “Everyone in the village is familiar with each other; there are few conflicts and disputes. If there are any problems, they come to us directly”.

Furthermore, the relational foundation may even lead villagers to support or shield grid workers when they face performance pressure. For instance, when uploading incident reports—which require the personal details of residents—one grid worker explained: “I usually just fill in their information myself. They understand—we all live in the same village and know each other” (DJ, ZL village grid worker). During interviews with villagers, even though they were unfamiliar with the grid system or “what is a grid worker,” they frequently expressed trust and support for village cadres, stating: “If we have any problems, we just call them directly.”

This seemingly mundane interaction reflects that rural digital governance is embedded in long-standing social ties, where mutual understanding and trust replace formal consent. Similar to the emotional labor of street-level bureaucrats (Zhang, 2025), their relational labor—often invisible in official reports—enables compliance with data requirements and the smooth navigation of governance expectations, serving as a critical infrastructure beneath the surface of digitalization.

Discussion

While the digital anklet system appears to enhance local administrative services and maintain social stability, our findings indicate that digital anklets have significantly increased the cost of governance for the state, placing a huge digital burden on frontline actors. Village cadres and grid workers expressed that they rarely have time off. “Every day is spent filling out forms, working overtime on weekends, and even at night. We are constantly dealing with layer upon layer of tasks” (CK, UN village cadre; DL, ZL village grid worker).

Under such pressure, the focus of governance shifts from effectiveness to performance for visibility. As Zhao (2023) suggests, the state rewards visibility over effectiveness, which leads to a perverse incentive structure—such as the assessment of “quantity” and “rate.” If frontline governance is effective, it means fewer incidents of conflict; however, this does not reflect the effectiveness of DGS. As one township leader noted, this practice is like “catching fish by draining the pond.” It vividly highlights that in the daily work of local governments, the focus is often on inflating visible statistical data to gain rewards or recognition, rather than addressing underlying issues or maintaining long-term sustainability. Amid high-intensity and repetitive daily tasks, frontline officials are increasingly engaged in visibility performances. During field research across 11 villages, participants—whether grid workers or village cadres—took photos to document our interviews: “Nowadays, we have to leave traces of everything we do” (WJ, ZJ village cadre-grid worker).

Such performances of visibility not only rely on the relational labor of grid workers but often come at the expense of data quality, undermining the very reliability of the information that digital governance systems claim to enhance. As a result, the state's investment in digital practices frequently fails to yield effective returns. As discussed earlier in the case of “environmental events,” through interviews with grid workers, environmental events are more prevalent because they are considered dispensable issues that allow grid workers to easily “create” and “resolve” reports. With the cancellation of the daily event reporting requirement in township S, the number of reported events has consequently decreased. The identification of inaccurate data and information not only affects the effectiveness of decision-making but also increases the cost of governance. During field research across 11 villages, it was observed that through multiple rounds of Humanity Environmental Improvement Actions, every village environment is indeed clean and tidy. However, this tidiness comes at a high administrative cost. For example, during my fieldwork in ZX village, six PWP workers were cleaning along a main road less than 800 m long. This suggests that the government must pay 4800 RMB in salaries per month. In traditional village governance, rural public affairs, such as sanitation, are often managed by the informal system of village rules and regulations, which rely on villagers’ self-discipline to maintain cleanliness. The refined management and services of the DGS have significantly increased the cost of frontline governance, making villages more dependent on resources and oversight from higher levels of government (Wong et al., 2020). However, these efforts often mask deeper structural contradictions that remain unresolved.

Furthermore, the regular monitoring of 27 categories of “key and sensitive groups” within the grid system raises further concerns. In contrast to studies that highlight how the classification and surveillance of vulnerable groups may inadvertently reinforce existing social hierarchies (Broomfield and Reutter, 2022; Dencik et al., 2019), China's paternalistic approach to problem-solving intensifies residents’ struggles for resources from higher-level authorities, generating a sense of imbalance within villages. The state has made significant investments to assist and protect vulnerable groups in rural areas. For instance, five-guarantee households comprise approximately 1% of the village population. Each “five-guarantee household” receives a monthly subsidy of 1100 RMB, which is a substantial income for rural elderly individuals. This has indirectly caused dissatisfaction among other elderly villagers. Ironically, the policies and financial investments made by the state to protect grassroots people in need have become the most significant source of conflict and dispute along the rural grid. Some villagers have complained through the 12345-citizen hotline because they were not selected as five-guarantee households. Complaints not only reduce the village grid worker's score, but the higher-level government will also hand the issue to the village. However, a participant noted that: “We cannot solve the problem because it does not meet the policy standards, but we still need to respond to the higher-level government” (WY, HZ village cadre-grid worker). Although only two village cadres and one villager expressed complaints about this conflict, we cannot overlook the potential risks posed by expanding the top-down administrative system. Current demographic trends suggest a future increase in the number of five-guarantee households, thereby placing escalating pressure on government welfare programs. As the state becomes more involved—with increased resources and funding allocated to rural welfare initiatives—it is unlikely that these measures will fully mitigate grassroots-level conflicts. Instead, such investments may foster psychological disparities, potentially triggering new social conflicts.

Conclusion

From the uncertain and formalistic macro-level “digital village” initiatives to the concrete implementation of “digital grid governance,” this study reveals the experimental, flexible, and iterative nature of China's local governments in advancing digital governance transformation. This process is shaped not only by local socioeconomic contexts and on-the-ground realities, but also by the power structure of China's bureaucratic system. Against this backdrop, local governments demonstrate a degree of strategic agency—responding to top-down mandates while also adapting implementation to local needs. Their actions reflect an ongoing negotiation between fulfilling central directives and navigating practical constraints at the grassroots.

In the context of rural China, digital governance often functions more as a symbolic performance of administrative achievement than as an effective tool for improving governance capacity. Unlike studies that frame digital governance as a cost-effective solution—particularly in Western contexts shaped by fiscal austerity—this study exposes both the visible and invisible costs of digitalization: the high financial and labor demands of data platform construction, the human effort required for data collection, and the invisible repetitive labor imposed on frontline actors under the constraints of digital systems.

Grid governance in urban areas is often understood as an extension of the existing bureaucratic apparatus into communities. It has been criticized for gradually suffocating grassroots autonomy—transforming governance into a process where individuals are administered rather than empowered to resolve their own problems (Mittelstaedt, 2022). However, this study finds that the functioning of digital systems is highly dependent on the relational labor of frontline officials. Embeddedness in rural social networks enables them to circumvent institutional rigidities and meet performance targets through informal practices and flexible negotiations. In rural China, their relational labor is not peripheral but foundational to governance. It enables township and village officials to mobilize cooperation, diffuse tensions, and navigate the gap between top-down mandates and local realities.

Moreover, much like in Western contexts where citizens are reframed as “users,” China's digital governance model lacks genuine civic participation. Citizens are often imagined not as active stakeholders but as sources of problems to be managed. In this context, digital systems are idealized as omnipotent tools that can discipline frontline actors and deliver services to passive users. This not only fosters public dependency on government services but also creates a sense of imbalance among villagers; it simultaneously increases frontline officials’ workload and psychological pressure while driving up the “invisible” governance costs.

The metaphor of “dancing with anklets” provides a dynamic perspective, drawing attention to the performative, interactive, and negotiated nature of policy implementation among multiple actors. Rather than a cold, rigid “digital iron cage,” rural digital governance operates through warm relational mediation, where frontline officials act as embedded data intermediaries. Their everyday relational work constitutes the affective infrastructure that sustains digital governance. This context highlights not only the social foundations of state power in local contexts but also how interpersonal relationships remain indispensable governance resources within digital systems. It is impossible to ignore that the increasing prevalence of digital anklets may ultimately forge a digital cage. As more digital anklets are implemented—whether for villages, local governments, or even the national government—all may eventually become trapped in the digital cage. Nevertheless, because County F is a national pilot area that may be promoted as a model for other regions, its experience may not fully represent the broader landscape of rural China.