Multilevel power structure and local compliance with government transparency mandates: evidence from China's environmental transparency reform

Abstract

Despite the ample literature on government transparency, our knowledge about how the vertical power structure of governments shapes local compliance with government transparency mandates is still limited. This study sets out to address this gap. Specifically, we investigate how the central government's environmental information disclosure (EID) signal and provincial governments’ conflicting signal of economic growth affect, independently and interactively, city governments’ compliance with central EID mandates in the center-province-city hierarchical structure in China. We argue that the central EID signal positively affects city compliance, while the provincial signal of economic growth reduces it. Moreover, the provincial signal of economic growth negatively moderates the impact of the central EID signal. Empirically, with a panel dataset for city-level governments from 2008 to 2018, we found robust evidence strongly supporting our theoretical hypotheses.

Points for practitioners

This research reveals the influences of the complex dynamics among governments at different levels on local compliance with government transparency mandates. The findings suggest that the maneuvers of middle-level governments in a multilevel power structure and the interactions among multiple conflicting policy goals should be taken seriously by practitioners when designing policies to promote government transparency reforms.

Keywords

China conflicting signals environmental transparency government transparency multilevel power structure

Introduction

Government transparency, which can be generally defined as the disclosure of information about governments’ workings or performance, among others, to the public at large (Grimmelikhuijsen and Feeney, 2017; Hood, 2007; Meijer, 2013) and is widely deemed to be “the key to better governance” (Hood, 2007),¹ has been increasingly emphasized across the globe during the past decades (Piña and Avellaneda, 2019). Meanwhile, as with other policy initiatives formulated at the central level, the actual effectiveness of the government transparency initiatives relies on the compliance at the local level. Therefore, understanding the underlying forces shaping local behaviors in information disclosure has become the focus of transparency research in public administration and political science.

Scholars have proposed various mechanisms that might influence the transparency of local governments. Generally, the perspectives adopted can be classified into two categories: the internal perspective that emphasizes the influences of organizational characteristics and the external perspective that focuses on the impacts of external pressures. Specifically, on the one hand, studies with the internal perspective suggest that a high-level organizational capacity (e.g. fiscal resources, technical expertise, professionalism, and leadership) and an internal organizational environment (organizational culture, ideology, structure, and organization managers’ attitudes towards transparency) conducive for transparency are crucial conditions for local governments to promote transparency (Bearfield and Bowman, 2016; Citro et al., 2021; Dragoş et al., 2012; Grimmelikhuijsen and Feeney, 2017; Sol, 2013). On the other hand, taking the external perspective, scholars highlight the influences of political competition (Bearfield and Bowman, 2016; Berliner, 2017) and pressures and demands from various external stakeholders, such as the general public, interest groups, and the civil society (Anderson et al., 2019; Bearfield and Bowman, 2016) in making local governments more (or less) transparent.

However, a notable gap in the extant literature is that hardly any research has investigated the influences of the complex vertical bureaucratic dynamics in a multilevel power structure on local governments’ compliance with transparency mandates. This scholarly omission is puzzling given that most countries in the world have a vertical power structure with multiple levels of authority, and the significant impacts of higher-level authorities on local policy behaviors have been broadly recognized in the public administration and political science literature. While recent studies have appealed for more attention to the role of superior governments in local transparency (Piña and Avellaneda, 2019), they tend to simplify the vertical interaction as a center–local relationship by assuming that the dynamics among superior authorities at different levels are negligible, and that the superiors can be regarded as one unitary actor. This assumption could be wrong. As widely noted in the literature, middle-level governments, as actors possessing their own interests, are not necessarily faithful agents of the central authority (Chen and Jia, 2022; Zhou, 2010; Zhu and Zhang, 2019). Therefore, it is both beneficial and necessary to go beyond this simplification and further probe how the complex vertical bureaucratic dynamics within a multilevel power structure affect government transparency at the local level.

Notably, this issue is even more prominent when studying government transparency in authoritarian contexts. While most of the extant literature on government transparency is conducted in western democratic countries, scholars have recently noted that authoritarian regimes, such as China, have adopted government transparency reforms to buttress their governance (Anderson et al., 2019; O’Connor et al., 2019; Wang, 2018; Xiao, 2013). Unlike their democratic counterparts, in authoritarian regimes wherein general elections are absent, local officials are mainly upwardly accountable to higher-tier political superiors. Therefore, while other mechanisms revealed in the democratic mechanisms might also be applicable, the influences of the vertical bureaucratic dynamics within a multilevel power structure are non-negligible and even more crucial in studying government transparency at the local level in authoritarian regimes.

This research sets out to fill this research gap. Taking China's environmental transparency (or environmental information disclosure, EID) reform as the case, we analyze how the vertical bureaucratic dynamics within the center–province–city hierarchical power structure affect city governments’ environmental transparency. Specifically, we investigate how central governments’ EID signal and provincial governments’ conflicting signal of economic growth signal affect, independently and interactively, city governments’ compliance with EID mandates through the multilevel power structure. We assert that central government’s EID signal positively affects cities’ compliance, while provincial governments’ conflicting signal of economic growth plays a negative role. Moreover, provincial governments’ conflicting signals of economic growth will attenuate or even neutralize the influences of central EID signals. To empirically validate the hypotheses, we obtained a panel dataset for city-level governments from 2008 to 2018 with data from various sources. Regression results find strong and robust evidence supporting our theoretical predictions.

This research contributes to the literature on government transparency at the local level by bringing in the complexities engendered by a multilevel power structure. Specifically, we reveal that the complex vertical bureaucratic dynamics among authorities at different levels will significantly shape local governments’ behaviors in promoting transparency. This study echoes the recent studies that emphasize the influences of higher-level governments (e.g. Piña and Avellaneda, 2019), but goes beyond them by further probing the impacts of the complex dynamics among higher-up authorities at different levels.

The remainder of the research is organized as follows. The next section reviews the literature on environmental transparency reform and local compliance in authoritarian China, which also provides basic information on the institutional context. The third section develops the theory on how the signals from the central and provincial governments affect city governments’ compliance with the EID mandates in China. The fourth section elucidates the research design, and the fifth section presents the empirical results. The sixth section provides a discussion, and the last section concludes the research.

Environmental transparency reform and local compliance in authoritarian China

China's transparency initiatives came from the top (Wang, 2018), and EID was part of the larger reform of “open government information”. The 2007 promulgation of Regulations on Open Government Information of the People's Republic of China (ROGI) and Measures on Environmental Open Information (MEOI), as the first sectoral operationalization of the ROGI (both took effect in May 2008), is widely regarded as a milestone of China's EID reform (Zhang et al., 2010). Both ROGI and MEOI imposed on local governments the obligation to disclose a broad range of environmental information, such as data on environmental quality, information on environmental management and supervision, and environmental laws and regulations (Wang, 2016). Over the follow-up years, the Chinese central government successively introduced various measures, amendments, and regulations to improve the EID rules. A notable one is the implementation of the New Environmental Protection Law in 2015, where “Information Disclosure and Public Participation” was listed for the first time as one of the only six chapters of the law (Wang, 2018).

Why does China, an authoritarian state traditionally more known for secrecy than transparency (Baekkeskov and Rubin, 2017), embrace environmental transparency reforms? This is an intriguing question that has attracted much scholarly attention. Generally, the extant literature tends to regard it as the result of the regime's governance transition. China has recently shifted away from the growth-at-all-costs path to one that emphasizes sustainable development (Chen, 2023). Since its eleventh five-year plan (2006–2010), environmental performance targets have been incorporated into the cadre evaluation system, indicating the center's substantial elevation of the priority of environmental protection (Wang, 2016). Against this background, the EID reforms are actually a tool adopted by the center to control local governments and thus improve environmental performance. EID, as a well-established informational environmental regulation instrument (Wang, 2018), is widely found to be conducive to improving local environmental performance (Tian et al., 2016). Specifically, the center in China, as with that in other authoritarian regimes, is plagued by the problems of asymmetric information and local noncompliance (Anderson et al., 2019). EID helps to rein in local governments by both mitigating vertical information asymmetry (Anderson et al., 2019) and generating bottom-up public pressure against local malfeasance as a complement to the traditional top-down command-and-control (Johnson, 2011; Lorentzen et al., 2014).

As for the local side, it is widely recognized that the level of EID, or the level of compliance with EID mandates, varies significantly across China's different local governments and different years (Tian et al., 2016; Zhang et al., 2010). A considerable thread of literature has thus emerged recently to investigate the underlying causes for the local variance in the compliance with EID mandates from various perspectives, which includes local leaders’ political incentives, local fiscal resources, and the influences of nonstate actors, among others. First, political incentives play a significant role in shaping local governments’ behaviors in EID. As with other environmental regulations, EID is largely in conflict with local economic development (Van der Kamp et al., 2017). Strict EID regulations impose restrictions on local industrial enterprises and might even force them to leave the current jurisdiction (Feng and He, 2020), which in turn hinders the local economy. In China, despite the fact that the center has incorporated environmental performance into the local cadre evaluation system since 2006, economic performance generally plays a more significant role in determining local officials’ political promotion (Chen and Jia, 2022; Kostka, 2016). Based on this perspective, previous literature found that local leaders’ various characteristics, such as tenure, associated with their eagerness to be promoted have significant impacts on their levels of EID (Tian et al., 2016). Second, implementing EID mandates requires considerable input of resources. Even if local governments have incentives to promote environmental transparency, they need enough fiscal resources to actually do it. Thus, local governments’ fiscal revenues are widely recognized to significantly affect their compliance with EID mandates (Van der Kamp et al., 2017). Third, besides local governments, the extant literature also underlies the influences of nonstate actors, such as non-governmental organizations (NGOs), the citizens, and industrial enterprises. On the one hand, the EID reform actually created additional space for civil society (Johnson, 2011), which in turn expanded the influences of actors such as NGOs and the general public on local environmental transparency (Anderson et al., 2019; Tan, 2014). On the other hand, as previously noted, EID adversely impacts local industrial enterprises on which local governments rely to achieve local economic development, among other goals. Local industrial enterprises, especially those giant ones, have many channels to capture local governments and undermine the implementation of EID at the local level (Lorentzen et al., 2014).

The fruitful research findings aside, the extant literature has largely neglected the influences of vertical bureaucratic dynamics on local governments’ compliance with EID mandates. This scholarly omission is puzzling, given the fact that China is an authoritarian state with a five-level governing pyramid from the center down to the township, and local governments’ implementation of mandated policies, EID mandates included, is widely recognized to be largely, if not mostly, affected by the complex and vertical bureaucratic dynamics within the multilevel power structure (Zhu and Zhang, 2019).

In summary, the environmental transparency reform in China provides researchers a great opportunity to investigate, both theoretically and empirically, how a multilevel power structure affects local compliance with government transparency mandates.

Theory and hypotheses

Focusing on the center–province–city hierarchical power structure, this section develops the theory on how the environmental transparency policy signal sent by the center and provincial governments’ conflicting signal of economic growth affect, both independently and interactively, city governments’ compliance with environmental transparency mandates in China.

The influence of the central policy signal of environmental transparency

The fundamental institution of China features a combination of political centralization and regional administrative decentralization (Xu, 2011). On the one hand, owing to the extensive decentralization reform since the 1980s, subnational governments possess considerable autonomy in making local administrative and policy decisions and the overall responsibility for their jurisdictional affairs (Chen et al., 2023). On the other hand, the center holds firmly the political and personnel power and retains substantial bureaucratic control over its local agents (Tsui and Wang, 2008). The results of this institution are twofold. First, local governments as self-contained actors are granted considerable room for maneuvering when implementing multiple mandates from the center. Second, owing to the centralized political power structure, local governments still have strong political incentives to respond to mandates that are deemed as the priority policies of the center. Central government's signal through official policy documents, among other tools, is a crucial mechanism that helps deliver the center's policy priority (Li, 2021) and thus plays a significant role in shaping local compliance. A strong signal of environmental transparency from the center indicates a high priority of EID in the central policy agenda, which in turn generates compliance pressures that are passed down level by level along the hierarchical administrative ladder. We thus derive the first hypothesis as follows:

Hypothesis 1:

Central government's signal of environmental transparency is positively associated with city governments’ compliance with the EID mandates.

When provincial governments issue a conflicting signal of economic growth

What happens when provincial governments, the middle-level superiors, issue a conflicting signal of economic growth? Essentially, in this situation, city governments, with limited attention and resources (Chen and Jia, 2021), are faced with the problem of prioritization between the two conflicting policies: economic growth and environmental transparency. To begin with, with the one-level-down management system (O’Brien and Li, 1999), provincial governments are the direct superiors possessing the full authority to appoint their city-level subordinates, among other authorities. City governments thus have strong political incentives to respond to their provincial superiors’ strong signal of economic growth and prioritize economic growth, which in turn negatively affects their compliance with central EID mandates.

Moreover, even without considering the direct vertical bureaucratic control of provincial governments, city governments per se have a strong tendency to prioritize economic growth over environmental transparency.

First, city governments in China have strong political and fiscal incentives to promote economic development (Xu, 2011). On the one hand, better economic performance compared with peers is widely believed and found to be significantly related to the political promotion of local officials at the sub-provincial level in China (Landry et al., 2018). On the other hand, the Chinese central government's strict control over local expenditure and local governments’ overall responsibilities for local affairs owing to regional decentralization also creates significant pressures on local governments to promote economic development and generate revenue (Van der Kamp et al., 2017).

On the contrary, except for the vertical pressures of EID mandates, city governments’ incentives to promote environmental transparency are limited, if not non-existent. In fact, owing to the characteristics of EID in authoritarian regimes, city governments possess strong motivations to resist environmental transparency. First, the negative impact of EID on local economic growth conflicts with local interests because economic development is crucial for local officials’ political promotion and city governments’ ability to fulfill other policy targets, such as providing public goods, among others (Edin, 1998). Second, as previously noted, the main goal of EID reform in China is a tool adopted by the center to enhance its control over its local agents (Anderson et al., 2019). Local governments as agents that tend to oppose anything that helps the principal to monitor and control them (Zhu et al., 2022) thus have incentives to resist the EID mandates. Third, local governments in China are vigilant about EID because it might undermine the policy objective of maintaining jurisdictional social stability (Wang, 2018). Maintaining societal stability is the priority goal for local governments in China with veto power (Wang, 2015). Failure in this goal would result in a veto for the appraisal of the overall performance of local leaders and even severe sanctions from the political superiors. EID, as with other types of government transparency, enhances civic engagement (Baekkeskov and Rubin, 2017; Johnson, 2011) and serves to coordinate citizen beliefs (Hollyer et al., 2015), which facilitates the emergence of collective action, impairing eventually jurisdictional social stability.

Therefore, a strong signal of economic growth from provincial governments, the middle-level superiors, is compatible with city governments’ own interests and provides them with additional “legitimacy” to prioritize economic growth and defy the EID mandates.

In summary, when a provincial government issues a strong signal of economic growth, its city agents, either out of vertical bureaucratic control of the provincial superior or their own interests, are more inclined to prioritize economic development. Given the conflicting relationship between economic development and environmental transparency, as well as city governments’ limited attention and resources, provincial governments’ conflicting signals of economic growth will cause a lower level of city compliance with the EID mandates and attenuate the impact of central governments EID signals. We thus have the second set of hypotheses as follows:

Hypothesis 2a:

Provincial government's conflicting signal of economic growth is negatively associated with city governments’ compliance with the EID mandates.

Hypothesis 2b:

Provincial government's conflicting signal of economic growth negatively moderate the impact of the central government's EID signal on city governments’ compliance with environmental transparency mandates.

Research design

Variables, measures, and data sources

Dependent variable: City governments’ compliance with EID mandates

This research uses the Pollution Information Transparency Index (PITI), an authoritative and objective measure that has been widely applied for local governments’ compliance with the central authority's EID mandate in China during the past decade (Anderson et al., 2019; Lorentzen et al., 2014; Van der Kamp et al., 2017), as the dependent variable. In 2009, two NGOs, the Institute for Public and Environmental Affairs (IPE) and the Natural Resource Defense Council (NRDC), created the PITI scoring system for 113 cities as an attempt to rate and monitor local governments’ compliance with environmental transparency. Corresponding to the ROGI, MEOI, and the successive flurry of central regulations, the PITI is designed to include a comprehensive set of subindices (e.g. the disclosure of environmental monitoring, information on environmental impact assessment, information on firms’ pollution violations and fines extracted on polluters, and responsiveness to public requests and complaints)² that reflect the requirements of the central EID mandates. Since 2009, IPE and NRDC have conducted evaluations and released PITI reports annually, which allows us to compile a panel dataset to examine local governments’ policy behaviors in EID. The various subindices are weighted and aggregated to a score ranging from 0 to 100. A higher PITI score indicates a city's better performance in EID. Notably, because the PITI is conducted by NGOs, it removes the self-reporting problem plaguing the traditional top-down target responsibility system and is thus regarded as more reliable data for assessing local governments’ compliance with EID (Anderson et al., 2019).

It is also worth noting that in the 2013–2014 report, IPE and NDRC modified significantly the compiling method (the subindices and their weights) of the PITI according to the newly issued EID-related central regulations (laws) and extended the city sample by including seven additional cities. However, because the modified PITI since 2013 actually reflects city governments’ compliance with the “new” EID mandates, which imposed different requirements on local environmental transparency, it still validly measures our dependent variable, that is, local compliance with central EID mandates. In this sense, while the compiling method of the PITI was modified in 2013, it is still reasonable to include the observations before and after 2013 together in the sample because they both validly reflect local compliance with the specific central EID mandates within the period that they cover. Four provincial-level cities (Beijing, Tianjin, Shanghai, and Chongqing) are excluded from the sample because they are directly controlled by the center and do not have provincial superiors. The final dataset thus comprises city-level observations (the numbers of cities are 109 and 116 in the pre- and post-2013 periods, respectively) from 2008 to 2018³ (the year that the latest PITI report covers).⁴

Key explanatory variables

Corresponding to the hypotheses, two key explanatory variables are included to measure central government's signals of environmental transparency and provincial governments’ conflicting signals of economic growth.

First, central governments’ EID signals are measured by the number of times that the central government directly requires local governments to disclose pollution information in the official regulatory documents issued in a year. Regulatory documents are unified guidelines issued by the central government to highlight the governance intentions, imposing clear requirements on local governments’ behaviors, and are considered to play a crucial role in China's governance system (Li, 2021; Xiao and Zhu, 2022). To obtain a complete collection of EID-related documents issued by the central government, we searched the PKU law database (www.pkulaw.com), which is widely used by scholars to systematically collect policy documents in a wide range of policy areas in China (Li, 2021), with the broad keywords of “environmental protection/governance” and “environmental pollution”. We then manually selected documents that contain requirements on EID and retrieved 411 relevant central documents in total for the period of 2008–2018. We performed a content analysis on these documents and calculated the number of times that the central government mentioned EID in the documents released in a certain year as the measurement of the central EID signal in that year.⁵ This variable is log-transformed to avoid potential biases caused by the skewed distribution and extreme values, among others, and thus obtain more valid estimations (West, 2022). We also use the number of relevant central documents released in a certain year (log-transformed) as an alternative measurement in the robustness checks.

Second, the provincial governments’ signal of economic growth is measured by the gap between economic growth targets set by provinces and the center in their annual government work reports (GWRs). In China, GWRs have long been a highly institutionalized system for higher-level governments to convey policy priorities to their subordinates (Chen and Jia, 2021). As the highlight of the annual “Two Sessions” (National People's Congress and the Chinese people's Political Consultative Conference) at the beginning of the year, the central and local governments at various levels will promulgate their GWRs to outline goals and development plans for the current year (Ma, 2016). Specifically, GWRs of governments at different levels adopt a largely uniform format with a particular “core” paragraph to highlight specific targets for different policy goals, and nearly all the GWRs will include the economic growth target. The targets specified in the GWRs represent serious political commitments to the public and the delegates, making the economic target a credible signal of economic growth for the city-level subordinates. Provincial governments, owing to the administrative decentralization regime, have substantial autonomy in making jurisdictional administrative and policy decisions, and economic growth is among them. Notably, provinces are widely found to defy central instructions in economic growth, particularly when the center attempts to control the overheated economy (Ma, 2016). We thus infer that a higher province–center discrepancy in the economic growth target conveys a stronger signal of provincial governments’ priority for economic growth (Du and Yi, 2022). To obtain the data on this measurement, we manually collected GWRs of the central and provincial governments in the period of 2008–2018 and coded the economic targets specified in these GWRs. The measurement for a certain province is then calculated by subtracting the central economic growth target from the province's own target for economic growth. We also use the absolute provincial economic targets as an alternative measurement in the robustness checks.

Control variables

This study includes a rich set of variables at both the city level and the provincial level to control for potential confounding impacts. First, we control for various city-level socioeconomic variables that are widely used in the previous literature (Lorentzen et al., 2014; Tian et al., 2016; Van der Kamp et al., 2017). These variables are GDP per capita (logged), GDP growth rate, the share of the secondary industry in GDP, the number of large industrial enterprises (logged), and population density (logged). The data for these variables were acquired from the China City Statistical Yearbook.

Second, we use a set of provincial level variables to control for the economic development statuses of the provinces, which include GDP per capita (logged) and GDP growth rate. Including these variables helps avoid/mitigate the potential biases caused by omitted variables. Specifically, it is plausible that more developed provincial governments tend to set higher economic growth targets and, at the same time, attend less to environmental governance and, thus, EID. This mechanism may also generate the empirical pattern of a negative relationship between provincial governments’ economic targets and city governments’ PITI scores.⁶ We obtained the data for these variables from the China Statistical Yearbook.

Descriptive statistics for dependent, explanatory, and control variables are presented in the appendices (Part A).

Estimation strategy

Considering the longitudinal nature of the dataset, this study adopts a standard linear panel data approach with the ordinary least squares (OLS) estimator to perform the analyses. The model specification corresponding to Hypothesis 1 and Hypothesis 2a is as follows:

P I T I_{i, t} = α + β_{1} E I D_C_{t} + β_{2} G r o w t h_P_{i, p, t} + λ X_{i, t} + γ_{i} + ε_{i, t}

(1)

where

P I T I_{i, t}

is the PITI score for city i in year t .

α

is the intercept.

E I D_C_{t}

represents the central EID signal in year t .

G r o w t h_P_{i, p, t}

denotes the signal of economic growth issued by city i ’s provincial superior p .

X_{i, t}

refers to a vector of control variables.

γ_{i}

denotes city fixed effect that eliminates confounding impacts of time-invariant factors.

ε_{i, t}

represents standard errors clustered at the city level. Of note, we do not control the year fixed effect because including year dummies leads to the perfect collinearity with

E I D_C_{t}

, our key explanatory variable that varies only over time.

To test the moderating effect illustrated in Hypothesis 2b, we further add an interaction term of central EID signals and provincial signals of economic growth signals into equation (1) and obtain the following econometric model:

P I T I_{i, t} = α + β_{1} E I D_C_{t} + β_{2} G r o w t h_P_{i, p, t} + β_{1} E I D_C_{t} \times G r o w t h_P_{i, p, t} + λ X_{i, t} + γ_{i} + ε_{i, t}

(2)

Empirical findings

Main regression results

As presented in Table 1, four models are estimated using standard fixed-effect panel data OLS regressions to provide the main regressions results.

Table 1.

Panel data regression results.

	Model 1	Model 2	Model 3	Model 4
Explanatory variables
Central EID signal (logged)	5.415*** (0.386)	4.659*** (0.623)	3.332*** (0.501)	2.140*** (0.609)
Province-center growth target discrepancies	−1.174*** (0.367)	−0.971*** (0.384)	−1.430*** (0.397)	−2.021*** (0.462)
Central EID signal(logged) × Province-center growth target discrepancies				−0.974*** (0.344)
City-level control variables
GDP per capita (logged)		1.011 (2.279)	−9.212*** (2.966)	−9.093*** (3.003)
GDP growth rate		−0.123 (0.139)	0.145 (0.159)	0.119 (0.161)
Number of large industrial enterprises (logged)		4.253** (1.778)	4.567** (1.894)	6.034*** (1.951)
Share of secondary industry in GDP		−0.185* (0.094)	−0.066 (0.101)	−0.023 (0.104)
Population density (logged)		1.003 (1.006)	0.885 (1.003)	0.695 (1.003)
Province-level control variables
GDP per capita (logged)			19.460*** (3.292)	22.242*** (3.217)
GDP growth rate			0.063 (0.321)	0.112 (0.324)
Within R-squared	0.391	0.399	0.425	0.431
Number of observations	1132	1126	1126	1126
City fixed effect	YES	YES	YES	YES

Standard errors clustered at the city level are reported in the parentheses. ***p < 0.01, **p < 0.05; *p < 0.1. The numbers of units (cities) are 109 and 116, respectively, for the 2008–2012 and 2013–2018 periods. The number of observations drops by six when including city-level control variables owing to missing data. In Model 4, the variables of central EID signal and province–center growth target discrepancies are mean-centered in the interaction term to ensure that interpretation of the coefficients for these two variables themselves is consistent with that in other models.

Model 1 offers a baseline estimation excluding the city-level and province-level control variables. The result shows that the estimated coefficients for the central EID signals and provincial economic growth signals are both statistically significant (at the 0.01 level), indicating that the central and provincial superiors have independent and significant influences on their city-level subordinates’ compliance with EID mandates. Specifically, as expected, a stronger central EDI signal contributes to a higher level of city governments’ compliance with EID mandates, and a stronger conflicting signal of economic growth from the provincial superiors impedes local compliance. Model 2 incorporates the city-level control variables, and Model 3 further includes the province-level control variables. The results for both models are quite robust for the two key explanatory variables. The findings of the above three models lend strong support to Hypothesis 1 and Hypothesis 2a, implying the independent influences exerted by superiors at different levels on lower subordinates’ compliance with environmental transparency mandates within a multilevel power structure.

Corresponding to Hypothesis 2b, Model 4, using the econometric model specified in equation (2), further analyzes how the moderating effect of provincial governments’ conflicting signal of economic growth on the impacts of central EID signals. The results show that the estimated coefficient for the interaction term is negative and statistically significant (at the 0.01 level), indicating that the impacts of the central EDI signal on local compliance with the EID mandates will be attenuated when provincial governments, the middle-level superiors, issue a strong conflicting signal of economic growth.

Figure 1 visually presents the above moderating effect. Specifically, the figure shows that the impacts of the central EID signal on city compliance decline as the province–center discrepancy in economic growth target gets larger. Notably, the impact of the central EID signal becomes insignificant when the discrepancy is considerably large. This indicates that when the provincial governments issue a conflicting signal that is strong enough, the influence of the central government's EID signal can be even fully neutralized. These results strongly support Hypothesis 2b.

Figure 1.

Marginal effects of central EID signals on city PITI scores (with 95% confidence intervals).

Robustness checks

Several robustness checks were also performed to further validate the main empirical results, including regressions with standard errors clustered at a higher level (the province level), regressions using hierarchical linear models, regressions with the 2013–2018 subsample, and regressions using alternative measurements of key explanatory variables. The results of these additional analyses confirm the robustness of the main regression results. Owing to the space limitation of this research, the results of the robustness checks are provided in the appendices.

Discussion

The findings of this research imply a challenge to China's environmental transparency reform when considering the complex bureaucratic dynamics within its multilevel power structure. That is, the provincial governments, as the middle-level superiors, possess the ability and, very likely, the intention to undermine the center's initiative of EID reform. When the center issues EID policy documents, owing to the centralized political power, the provinces have to follow by either conveying the documents to their city subordinates or issuing their own version of the policies according to the central guidance. Given provincial governments’ strong control over their city subordinates (Mertha, 2005), the center has the ability to convey its EID mandates down to the cities to promote their compliance. However, as found in this study, the provinces can easily and lawfully attenuate or even neutralize the influences of the center by sending a strong signal of economic growth without explicitly defying the center's EID mandates. Considering that the provinces, as the agents of the center, also have the inclination to resist the center's control via the EID reform and the incentives to promote economic growth, we believe that the provincial governments also possess the intention to weaken the influences of central EID mandates. This might provide another explanation for the relative slowness of the overall progress in EID at the local level, according to the PITI reports. Notably, in this sense, our study also speaks to the bureaucratic control literature (e.g. Xiao and Zhu, 2022) by revealing how the center's control, albeit with highly centralized political power, can be jeopardized by the maneuvers of middle-level governments in a multilevel power structure. In summary, bearing this case in mind, we can confirm again the importance of bringing the multilevel-power-structure perspective into the research on local government transparency, and beyond.

Also, we would like to elaborate on the external validity and, relatedly, the contribution of this research. Obviously, the specific theoretical predictions of this study are based on the context of China's environmental transparency reform and might not be directly applicable to other contexts. However, in a broader sense, what this research suggests is the importance of going beyond the simplification of regarding the superior authority as one unitary actor and further probing the complexities engendered by the vertical bureaucratic dynamics among superior authorities at different levels. The multilevel power structure is adopted by most of countries around the world. While the specific institutional settings (e.g. the distribution of political and administrative powers among governments at different levels) might be different, we believe the general idea that different signals sent by superior authorities at different levels affect, independently and interactively, local governments’ compliance with government transparency mandates is generally applicable to other contexts, no matter if it is authoritarian or democratic. In this sense, this research, at the very least, makes a contribution to the literature on government transparency at the local level by introducing a new and crucial perspective to study the vertical influences of higher-up authorities.

Conclusion

Despite the high volume of literature on government transparency at the local level, little attention has been paid to investigating how the vertical bureaucratic dynamics among superior authorities at different levels affect local compliance with transparency mandates. This research attempts to fill this gap by studying how the signals from the central and provincial governments influence, independently and interactively, city governments’ compliance with the central EID mandates in China. We find that while central EID signals increase local compliance, provincial governments’ conflicting signals of economic growth decrease it and, moreover, attenuate the influence of central EID signals.

In a broad sense, this research echoes the recent appeal for more scholarly attention to the influences of higher-up governments when studying local government transparency (Piña and Avellaneda, 2019) and goes further by investigating how the bureaucratic dynamics among superior authorities at different levels shape local compliance with government transparency mandates in a multilevel power structure. The findings of this research have important implications for understanding government transparency in China, as well as other regimes with a multilevel power structure.

This research only represents an initial step forward for a broader research agenda. Several limitations and potential avenues for future research are worth noting. First, we only study a situation wherein the signals from superiors at different levels are conflicting. Future research can further explore what will happen when the signals are complementary. Second, owing to the data limitation, we study the dynamics in a three-level center–province–city structure. It causes the empirical caveat that we cannot control the year fixed effect because the signals of the highest authority (the center) vary only over time. It would be beneficial to use lower-level governments as the unit of analysis when data is available. Third, our study is based on the case of China’ environmental transparency reform. Future research can extend this research by studying other sorts of government transparency (e.g. fiscal transparency) or generating evidence from other regimes.

Supplemental Material

sj-docx-1-ras-10.1177_00208523231167081 - Supplemental material for Multilevel power structure and local compliance with government transparency mandates: evidence from China's environmental transparency reform

Supplemental material, sj-docx-1-ras-10.1177_00208523231167081 for Multilevel power structure and local compliance with government transparency mandates: evidence from China's environmental transparency reform by Jin Li and Shaowei Chen in International Review of Administrative Sciences

Footnotes

Authors’ contribution

The authors contributed equally to this work and should each be considered as co-first author.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This work was supported by the National Natural Science Foundation of China (Grant 72104260) and National Social Science Foundation of China (Grant 18CGL036).

ORCID iDs

Jin Li

Shaowei Chen

Supplemental material

Supplemental material for this article is available online at:

Notes

Jin Li is Assistant Professor at the School of Government, Central University of Finance and Economics, China. Her main research interests are in environmental policy and politics, central–local government relations in China, and collaborative governance.

Shaowei Chen is Professor at the School of Public Administration, Hunan University, China. His main research interests are in environmental policy and politics, performance management, and central–local government relations in China.

References

Anderson

Buntaine

Liu

, et al. (2019) Non-governmental monitoring of local governments increases compliance with central mandates: A national-scale field experiment in China. American Journal of Political Science 63(3): 626–643.

Baekkeskov

Rubin

(2017) Information dilemmas and blame-avoidance strategies: From secrecy to lightning rods in Chinese health crises. Governance 30(3): 425–443.

Bearfield

Bowman

AOM

(2016) Can you find it on the web? An assessment of municipal e-government transparency. The American Review of Public Administration 47(2): 172–188.

Berliner

(2017) Sunlight or window dressing? Local government compliance with South Africa's promotion of access to information act. Governance 30(4): 641–661.

Chen

(2023) Performance feedback, blame avoidance, and data manipulation in the public sector: Evidence from China's official city air quality ranking. Public Management Review: 1–21. Epub ahead of print. DOI: 10.1080/14719037.2023.2171098.

Chen

Jia

(2021) How local governments prioritize multiple conflicting goals: Beyond the sole-goal perspective. Public Administration. Epub ahead of print. DOI: 10.1111/padm.12807.

Chen

Jia

(2022) Bringing in horizontal strategic interactions: Local governments’ bandwagon strategy and its implications for performance management. International Public Management Journal 25(4): 600–619.

Chen

Zhao

Zhou

(2023) Which works better? Comparing the environmental outcomes of different forms of intergovernmental collaboration in China's air pollution control. Journal of Environmental Policy & Planning 25(1): 115–134.

Citro

Cuadrado-Ballesteros

Bisogno

(2021) Explaining budget transparency through political factors. International Review of Administrative Sciences 87(1): 115–134.

10.

Dragoş

Neamtţu

Cobârzan

(2012) Procedural transparency in rural Romania: Linking implementation with administrative capacity? International Review of Administrative Sciences 78(1): 134–157.

11.

(2022) Target-setting, political incentives, and the tricky trade-off between economic development and environmental protection. Public Administration 100(4): 923–941.

12.

Edin

(1998) Why do Chinese local cadres promote growth? Institutional incentives and constraints of local cadres. Forum for Development Studies 25(1): 97–127.

13.

Feng

(2020) The effect of environmental information disclosure on environmental quality: Evidence from Chinese cities. Journal of Cleaner Production 276: 1–13.

14.

Grimmelikhuijsen

Feeney

(2017) Developing and testing an integrative framework for open government adoption in local governments. Public Administration Review 77(4): 579–590.

15.

Hollyer

Rosendorff

Vreeland

(2015) Transparency, protest, and autocratic instability. American Political Science Review 109(4): 764–784.

16.

Hood

(2007) What happens when transparency meets blame-avoidance? Public Management Review 9(2): 191–210.

17.

Johnson

(2011) Environmental information disclosure in China: Policy developments and NGO responses. Policy & Politics 39(3): 399–416.

18.

Kostka

(2016) Command without control: The case of China's environmental target system. Regulation & Governance 10(1): 58–74.

19.

Landry

Lü

Duan

(2018) Does performance matter? Evaluating political selection along the Chinese administrative ladder. Comparative Political Studies 51(8): 1074–1105.

20.

(2021) Signalling compliance: An explanation of the intermittent green policy implementation gap in China. Local Government Studies: 1–27. Epub ahead of print. DOI: 10.1080/03003930.2021.1997743.

21.

Lorentzen

Landry

Yasuda

(2014) Undermining authoritarian innovation: The power of China’s industrial giants. The Journal of Politics 76(1): 182–194.

22.

(2016) Performance feedback, government goal-setting and aspiration level adaptation: Evidence from Chinese provinces. Public Administration 94(2): 452–471.

23.

Meijer

(2013) Understanding the complex dynamics of transparency. Public Administration Review 73(3): 429–439.

24.

Mertha

(2005) China's ‘soft’ centralization: Shifting tiao/kuai authority relations. The China Quarterly 184: 791–810.

25.

O'Brien

(1999) Selective policy implementation in rural China. Comparative Politics 31(2): 437–460.

26.

O’Connor

Janenova

Knox

(2019) Open government in authoritarian regimes. International Review of Public Policy 1(1): 65–82.

27.

Piña

Avellaneda

(2019) Central government strategies to promote local governments’ transparency: Guidance or enforcement? Public Performance & Management Review 42(2): 357–382.

28.

Sol

DAD

(2013) The institutional, economic and social determinants of local government transparency. Journal of Economic Policy Reform 16(1): 90–107.

29.

Tan

(2014) Transparency without democracy: The unexpected effects of China's environmental disclosure policy. Governance 27(1): 37–62.

30.

Tian

Guo

Han

, et al. (2016) Different extent of environmental information disclosure across Chinese cities: Contributing factors and correlation with local pollution. Global Environmental Change 39(4): 244–257.

31.

Tsui

Wang

(2008) Decentralization with political trump: Vertical control, local accountability and regional disparities in China. China Economic Review 19(1): 18–31.

32.

Van der Kamp

Lorentzen

Mattingly

(2017) Racing to the bottom or to the top? Decentralization, revenue pressures, and governance reform in China. World Development 95(7): 164–176.

33.

Wang

(2018) Explaining environmental information disclosure in China. Ecology Law Quarterly 44(4): 865–923.

34.

Wang

(2016) Managing social stability: The perspective of a local government in China. Journal of East Asian Studies 15(1): 1–25.

35.

Wang

(2015) Requests for environmental information disclosure in China: An understanding from legal mobilization and citizen activism. Journal of Contemporary China 25(98): 233–247.

36.

West

(2022) Best practice in statistics: The use of log transformation. Annals of Clinical Biochemistry: International Journal of Laboratory Medicine 59(3): 162–165.

37.

Xiao

Zhu

(2022) Bureaucratic control and strategic compliance: How do subnational governments implement central guidelines in China? Journal of Public Administration Research and Theory 32(2): 342–359.

38.

Xiao

(2013) Freedom of information reform in China: Information flow analysis. International Review of Administrative Sciences 79(4): 790–808.

39.

(2011) The fundamental institutions of China's reforms and development. Journal of Economic Literature 49(4): 1076–1151.

40.

Zhang

Mol

APJ

, et al. (2010) An implementation assessment of China's environmental information disclosure decree. Journal of Environmental Sciences 22(10): 1649–1656.

41.

Zhou

(2010) The institutional logic of collusion among local governments in China. Modern China 36(1): 47–78.

42.

Zhu

Zhang

(2019) Diffusion of marketization innovation with administrative centralization in a multilevel system: Evidence from China. Journal of Public Administration Research and Theory 29(1): 133–150.

43.

Zhu

Qiu

Liu

(2022) Resisting public monitoring in authoritarian regimes: Evidence from local environmental litigation in China. Governance. Epub ahead of print. DOI: 10.1111/gove.12675.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

4.49 MB