Does vertical administrative integration of environmental monitoring at the provincial level effectively motivate firms to improve environmental behaviour in China?

Abstract

A few cities and provinces in China have implemented vertical administrative integration of environmental monitoring to the provincial level as a response to severe environmental pollution. This study used an adaptive agent-based simulation model to explore whether the reform might effectively motivate polluting industrial firms to improve their environmental behaviour. Simulation results found that the reform might not effectively motivate the desired improvements in environmental behaviour unless policy-makers improve individual enterprises’ financial capacities, enhance their subsidies, and encourage managers to improve their environmental awareness. These findings could be used in the vertical administrative reform efforts to help achieve the reform’s success.

Points for practitioners

The vertical reform needs to be sufficiently systematic across its governmental structure because it cannot operate in isolation. It is a part of the country’s complex economic, social, and environmental societal system. Combining administrative restructuring with regulation of micro-agents’ behaviour might increase the reform’s likelihood of success, and financial policies might improve preventive/enthusiastic environmental behaviour. A sophisticated policy approach, such as encouraging preventive/enthusiastic environmental behaviour through business opportunities, might ease behavioural change.

Keywords

agent-based model environment monitoring policy performance vertical administrative reform

Introduction

A key challenge to the management of the natural environment is achieving internal managerial and organizational alignment across levels as a way to increase performance. This is particularly the case in China, whose industrial pollution challenges have been documented by many researchers (Liu, 2017). For many years, China’s central government employed a ‘tournament competition’ approach among municipal governments that focused on Gross Domestic Product (GDP) as the key criterion to assess municipal officials’ promotions (Jiao et al., 2011). Therefore, municipal officials developed their local industrial economies and had little incentive to protect the natural environment. Many established industries acted abusively towards the environment (Song et al., 2015). Thus, industrial enterprises are the main causes of the pollution problems (Liu and Ye, 2012).

Municipal governments particularly preferred large industrial enterprises, such as state-owned industries, because they made large contributions to municipal GDPs (Ma et al., 2015). These large enterprises developed close relationships with local governments, which improved their likelihood of obtaining governmental support, including relatively low environmental monitoring and evaluation standards. In some cases, municipal governments even helped to hide environmental pollution caused by these large industrial enterprises (Ma et al., 2015). Furthermore, although the environmental officials were authorized to regulate these abusive environmental activities, they had little incentive to do so because their promotion might be jeopardized (Chen, 2016). For example, municipal governments determined their municipal environmental officials’ promotions.

In 2016, to strengthen municipal environmental officials’ powers to motivate industries to improve their environmental behaviour, the Chinese central government changed the management system by enacting the Environmental Vertical Management Reform (EVMR), which mandated that provincial governments determine the promotions of municipal and county (district) environmental officials. However, the extent to which the EVMR effectively motivates polluting enterprises to improve their environmental behaviour by weakening their power over municipalities is not clear. Additionally, the factors that influence behavioural changes have not been fully identified. Previous studies have partially explored these questions.

Vertical management is a system in which an agency works through an internal hierarchical frame, with lower sections reporting directly to upper ones instead of to outside local governments (Li, 2016). During its long history, scholars in many disciplines have studied vertical management. For example, vertical fiscal externalities (Andersson et al., 2004), airline alliances (Grauberger and Kimms, 2015), manufacturing (Carnovale et al., 2017), inventory management (Wan, 2017), and market competition (Loertscher and Reisinger, 2014) are widely covered topics. However, the known effects of vertical integration are open to widely differing interpretations (Jabes et al., 1992; Liddle, 2018; Uddin et al., 2019; Wescott, 2009), and three major streams of research have developed.

The first research stream has focused on the positive aspects of strategic vertical alignment. For example, according to Moore (1995), strategic vertical alignment might create public value, which was partly supported by Lane and Wallis (2009). The efficacy of vertical strategies was also considered important in establishing organizational strategies (Bowman and Ambrosini, 1997), particularly in the Spanish experience, where vertical competition mechanisms were likely responsible for policy innovations (Costa-Font and Rico, 2006). Furthermore, many researchers have offered convincing reasons for vertical coordination in food supply management, such as governmental regulation of product quality (Caswell et al., 1994), pesticide residue (Kilmer et al., 2001), and consumer awareness of food safety (Hennessy, 1996).

The second research stream has focused on the negative effects of vertical management, arguing that vertical structuring undermines political control because of weakened structural levers of control (Christensen and Lægreid, 2008). Mussa et al. (2013) found that a vertical funding system created inequalities in Mozambique’s national health system, which partly supported Unger et al. (2003). On the other hand, vertical systems might be coordinated as parallel yet separate systems to help improve the availability of information to decision-makers, although this benefit is often not realized (Kawonga et al., 2012; World Health Organization, 2008). Regarding vertical coordination, critics of Germany’s 2002 National Sustainability Strategy conclude that the performance of the national government must improve because some objectives might be reached only through collaboration with state and local agencies (Niestroy, 2005).

Another interesting research stream has focused on factors that influence the vertical integration of administrations. Organized transaction costs that minimize responses to opportunism, bounded rationality, and information impactedness might lead to vertical integration (Medema, 1992). Other factors, such as knowledge (Kawonga et al., 2013), collaboration (Sumerall, 1995), the direction of vertical integration (upward or downward) (Wang et al., 2001), and regulatory restrictions (Nunan et al., 2012; Vita, 2000), also have been found to influence the performance of vertically integrated administrations. Vertical integration is not always straightforward, it has different meanings for different people (Hopkins, 2009), and it has individual market-oriented strategic behaviour (Mick et al., 1993), which remain open to various explanations.

Therefore, vertical administrative integration has been studied, but to a limited extent, regarding positive and negative causes and effects, particularly the relationship between micro-level behaviour as outcomes of vertical administrative integration. This latter relationship was this study’s focus and its theoretical contribution because it examined the influence of vertical administrative integration on environmental behaviour. There also seems to be a lack of empirical studies on the relationship between vertical administrative integration with respect to environmental monitoring and industrial environmental behaviour in China (and elsewhere). This study focused on environmental monitoring, which is its empirical contribution to the literature. This study used an agent-based model to examine the role of the EVMR in motivating polluting enterprises to improve their environmental behaviour. It also considered factors that influence the relationship. The results provide rich data on vertical integration of administrations and for public policy-making.

Methods

The following two research questions were addressed using simulation methods.

Does vertical administrative integration of environmental monitoring to the provincial level effectively motivate enterprises to improve their environment-related behaviour?

What are the main factors influencing such behavioural changes?

Thus, the study’s theoretical framework was the basis of its conceptual model, and the simulation model analysed was designed using the conceptual model. The validity of these methods was theoretically and empirically addressed through the analysis described below.

Complex multi-agent model: why has it been employed?

Practitioners and scholars of public administration have developed a theory that public policy-making and public administration are complex processes (Teisman and Klijn, 2008). The ‘complexity theory’, which includes dynamism, non-linearity, and fitness as well as self-organizing agents, might help to increase the number of studies on change to public administration. Understanding the friendliness of the complexity theory to other relevant theories might facilitate the application of complexity theory to studies on public administration and extend the explanatory power of compatible theories (Eppel, 2017). The agents are self-organizing, and they self-create their perceptions of their desires and activities, although governments and public managers often seem to act as self-referential agents.

By employing complexity theory and agent-based models, many researchers have successfully analysed public administration problems. For example, Bovaird (2008) argued that agents’ behaviour and strategies owed at least as much to emergent complex interactions in the policy system as they owed to the cognitive processes of any one agent, and, then, they conducted a case study of UK municipal governments. According to Haveri’s (2006) findings, complexity was a dominant characteristic of change in municipal government, which he applied to the complexity of limits placed on rational reform processes. Another empirical study concerned urban regeneration projects, to which a complexity perspective was applied to analyse and interpret public decision-making (Rhodes, 2008). An empirical study conducted in New Zealand (Eppel, 2012) also found that complexity theory improved the understanding of public management scholars and practitioners.

On the other hand, the relationship between vertical environmental management reform and industrial environmental behaviour is complex because it is a multi-agent system. Environmental regulations change in response to agents’ interactions (including industries and governments), which, in turn, influences the agents’ behaviour. Thus, an agent-based model is considered useful because it can simulate the complexity of the agents, the dynamic evolution of their behaviour, and their interactions beyond the scope of traditional methods.

Models and their relationships

This study’s theoretical framework regarding the relationships of agents to complexity was the foundation of its multi-agent model. The conceptual model was developed from this foundation to focus on the relevant actors in mainland China (enterprises and governments) and on the system of environmental administration in the country. The variables regarding the agents and the interaction among them were in the conceptual model, which was the basis of the simulation model. The main purpose of the simulation model was to design a computer program to simulate the conceptual model and address the two research questions (Figure 1).

Figure 1.

The theoretical framework and modelling.

Conceptual model

The conceptual model was based on the theoretical framework to create a common framework for designing the multi-agent simulation model. In mainland China, the central government includes the Central Committee of the Chinese Communist Party, the State Council, and the National People’s Congress, which is mainly responsible for environmental management. First, the Central Committee of the Chinese Communist Party proposes a national guideline on environmental management, and, then, the State Council formulates an implementation plan, which is followed by legislation drafted by the National People’s Congress. The Ministry of Environmental Protection is authorized to implement specific environmental policies. Higher levels of government usually meet their environmental responsibilities by delegating them to lower levels of government. Historically, national environmental responsibility was distributed among all of the provinces. Provincial-level environmental protections were under the jurisdiction of the provincial governments, the municipal governments administered the municipal-level environmental protections, and the counties (districts) administered environmental protections at that level (Li, 2011). However, after vertical reform, the counties (districts) were replaced by the dispatched agencies on environmental protection at the municipal level, which were, in turn, under provincial administration (Figure 2).

Figure 2.

Structure of environmental administration in mainland China.

In mainland China’s environmental management, the system mainly includes government and enterprises with varying environmental behaviour. This study chose variables to simulate the system that were justified by previous studies or through data derived from a field survey (Table 1). For example, the measure of ‘Municipal-county-favouritism’ was developed to measure the extent of municipal governmental support of enterprises that had close relationships to municipal governments or large state enterprises. After vertical reform, the municipal and county governments were unable to conduct environmental monitoring and evaluations, and, therefore, the effects of Municipal-county-favouritism on environmental behaviour were expected to decrease.

Table 1

Main variables used in the simulations with their descriptions and sources.

Variable	Description	Data source
Bankruptcy-firms	Number of bankrupted enterprises	Liu and Ye (2012)
New-firms	Number of new enterprises	Liu and Ye (2012)
Defensive-behaviour-firms	Number of enterprises with defensive environmental behaviour	Liu (2009)
Preventive-behaviour-firms	Number of enterprises with behaviour aimed at preventing environmental damage	Liu (2009)
Enthusiastic-behaviour-firms	Number of enterprises enthusiastic about environmental protection behaviour	Liu (2009)
Financial-capacity	Financial capacity of an enterprise	Guan et al. (2005); Murillo-Luna et al. (2011)
Municipal-county-favouritism	Municipal or county governmental support, including low environmental monitoring and evaluation standards for enterprises with close relationships to local governments	Ma et al. (2015)
Environmental-assessment-preparedness	Extent of preparedness of environmental assessment	Liu and Ye (2012)
Environmental-awareness-manager	Managers’ environmental awareness	Ajzen (2001); Sudarmadi et al. (2001)
Profit	The profit of an enterprise	Liu and Ye (2012)
Subsidy	Whether an enterprise receives a governmental subsidy	Liu and Cui (2011)
Standard-environmental-regulation	Governmental environmental regulation	Reijnders (2003); Robert et al. (2003)
x1	The minimum profit of New-firms	Liu (2009)
x2	The minimum profit of Defensive-behaviour-firms	Liu (2009)
x3	The minimum profit of Preventive-behaviour-firms	Liu (2009)

Figure 3 illustrates the processes of the conceptual model designed for this study’s simulations. Enterprises’ paths are expected to differ because of their different environmental behaviour, which are influenced by different factors. The conceptual model presents three types of environmental behaviour. According to Liu (2009), enterprises with defensive behaviour are reluctant to comply and try to delay or oppose governmental environmental regulations. Those with preventive behaviour integrate environmental management into their internal systems through various changes, such as establishing environmental management positions. Enterprises demonstrating enthusiastic behaviour seek financial and sustainable rates of return, and environmental management is fully integrated into their businesses and cultures. Numerous factors influence environmental behaviour, such as Municipal-county-favouritism, Financial-capacity, and Environmental-awareness-manager, which creates a deeply interrelated and complex system. On the other hand, Guan et al. (2005) found that environmental behaviour positively correlated with an enterprise’s profits. Thus, the conceptual model depicts the theoretical set of factors expected to influence enterprises’ environmental behaviour.

Figure 3.

Flowchart of the conceptual model’s processes.

Simulation model

A NetLogo computer program was designed based on the conceptual model. The NetLogo platform is a multi-agent programming language and modelling environment for simulating complex adaptive systems. Using the NetLogo simulation platform, this study’s model simulated whether vertical administrative integration of environmental monitoring at the provincial level effectively motivated enterprises to improve their environmental behaviour, and explored the factors that influenced behavioural change. In the simulation model, agents were considered with respect to fitness as profit-maximizing agents; therefore, the fitness of an enterprise was calculated based on its profits. Furthermore, enterprises were considered agents with adaptive traits influencing their fitness in response to governmental environmental regulations while they learned from the political environment and from influential factors, through which they adjusted their environmental behaviour.

Stochasticity was in the model through variables, such as the location of an enterprise, and it was also used to control for the effects of some variables because the focus of the simulation model was on whole-system behavioural change. The simulated world comprised patches and an observer could instruct all of the agents (i.e. the enterprises). An observer could also change the values of the variables (e.g. Municipal-county-favouritism). Therefore, it was possible to examine virtual associations between individual enterprises’ environmental behaviour and governmental regulations after vertical reform occurred. On the NetLogo simulation platform, controls are on the left side and the graphics window is on the right side, where the simulated world is visible (Figure 4).

Figure 4.

NetLogo simulation platform illustrating the simulation model.

Fujian Province was chosen as the subject of the vertical reform simulation experiment (Figure 5). Located on China’s southeast coast, Fujian is industrialized with 9214 industrial enterprises that contributed 41.65 percent to the 2015 GDP. In 2016, there were 485 state-owned industrial enterprises, with total assets 1.18 times that of the privately owned industrial enterprises (China Statistical Yearbook, 2016).

Figure 5.

The location of Fujian in China.

Industrial pollution is a serious problem in Fujian. For example, the amount of industrial wastewater discharge and waste gas emissions increased by 334.31 percent and 745.99 percent, respectively, between 1998 and 2011 (Fujian Statistical Yearbook, 1999, 2012). Furthermore, the number of emergency environmental accidents ranked Fujian in the top five provinces in 2015 (China Statistical Yearbook, 2016). Table 2 describes the data used to anchor the beginning of the simulation. In other words, these were the initial baseline values of the variables used in the simulation.

Table 2.

Overview of the baseline data on the main variables in the simulation.

Variable	Baseline values
Max-firms	Fujian Province had 9214 industrial enterprises in 2015 (China Statistical Yearbook, 2016)
Municipal-county-favouritism	Randomly controlled variables^a from 4, 5, and 6
Environmental-awareness-manager	Randomly controlled variables from 3 (high), 2 (middle), and 1 (low)
Financial-capacity	Randomly controlled variables from 3 (high), 2 (middle), and 1 (low)
Environmental-assessment-preparedness	Randomly controlled variables from 3 (best), 2 (better), and 1 (good)
Yearly-financial-aid-available	CNY 3.55 million estimated using Fujian Province data
Standard-environmental-regulation	Randomly controlled variables from 3 (strictest), 2 (stricter), and 1 (strict)
x1	CNY 3.00 million estimated using China Statistical Yearbook (2016) and field survey data^b
x2	CNY 2.25 million estimated using China Statistical Yearbook (2016) and field survey data
x3	CNY 13.69 million estimated using China Statistical Yearbook (2016) and field survey data

^aTo analyse behaviour under different scenarios, the effects of some important variables were manually controlled for, allowing for various experimental scenarios.

^bData derived from the Fujian field survey found that enterprise size significantly influenced environmental behaviour. Most relatively large firms (CNY 20 million income or more per year) demonstrated enthusiastic environmental behaviour, and most of the relatively small enterprises (less than CNY three million income per year) demonstrated defensive environmental behaviour. However, most mid-sized enterprises (yearly incomes between CNY three and CNY 20 million) demonstrated preventive environmental behaviour.

Comparative simulation: before and after vertical administrative reform

Three computational experiments simulated the values before and after vertical administrative reform occurred. The baseline and parametric values found in the experiments (named S_B1, S_A1, and S_A2) are presented in Table 3. Experiment S_B1 included Municipal-county-favouritism as a randomly controlled variable (4, 5, and 6). Financial-capacity was also a randomly controlled variable (1, 2, and 3). The other controlled variables were Environmental-assessment-preparedness, Standard-environmental-regulation, and Environmental-awareness-manager. Their values were manually controlled. However, Yearly-financial-aid-available (3.55) was estimated using Fujian Province data.

Table 3.

Parameters of the simulation scenarios.

Variable	Simulation	Scenario
Before vertical administrative reform	S_B1
Municipal-county-favouritism	[4,5,6]
Financial-capacity	[1,2,3]
Environmental-assessment-preparedness	[1,2]
Yearly-financial-aid-available	3.55
Standard-environmental-regulation	[1]
Environmental-awareness-manager	[2]
After vertical administrative reform	S_A1	S_A2
Municipal-county-favouritism	[2,3,4]	[2,3,4]
Financial-capacity	[1,2,3]	[3]
Environmental-assessment-preparedness	[2,3]	[2,3]
Yearly-financial-aid-available	3.55	7.10
Standard-environmental-regulation	[2]	[2]
Environmental-awareness-manager	[2]	[3]

Validation of the model

Balci (1998) pointed out that there is no generally accepted way to assess verification and validity. Nevertheless, the modelling approach was considered as model verification, validation, and calibration. Sustained by the literature and multiple sources of data including field survey data improved the model’s validation particularly determining whether the conceptual model was a reasonably accurate representation of the real world (Law and Kelton, 1991).

Detecting and removing errors from the simulation program, which helped to eliminate an incorrect application of the conceptual model, also guaranteed the simulation model’s validity. Furthermore, an iterative process, which adjusted the unmeasured or poorly characterized experimental parameters, was employed to program the extreme conditions and test behaviour sensitivity. The calibration process improved the fit between the model and the data. The results of these tests indicated that the author identified changes in the model’s dynamic behaviour patterns that resulted from changes in the values of particular parameters corresponding to available knowledge of the real world. Thus, simulating scenarios with the simulation model was an acceptable way to determine strategies for improving the administrative system’s performance.

Results and discussion

Number of firms with preventive or enthusiastic environmental behaviour increased slightly after vertical administrative reform

After vertical reform, municipal or county governmental support, including low environmental monitoring and evaluation standards for enterprises with close relationships with municipal governments, is expected to decrease because provincial environmental protections control the environmental monitoring and evaluations, which would result in strict environmental regulations. Therefore, the value of Municipal-county-favouritism was randomly changed to low levels (2, 3, and 4). On the other hand, the values of Environmental-assessment-preparedness (2, 3) and Standard-environmental-regulation (1) were changed to higher levels. The comparative results indicated that, during the simulated period, the number of Defensive-behaviour enterprises would decrease, and the numbers of Preventive-behaviour and Enthusiastic-behaviour enterprises would slightly increase (Figure 6).

Figure 6.

Comparative simulation results of the number of firms before and after vertical reform.

Tests of homogeneity of variances, Levene’s test, analysis of variance (ANOVA), and F-tests were performed. The Levene’s test results yielded p-values greater than 0.05, and, therefore, ANOVA was an appropriate test. The ANOVA test result also had a p-value greater than 0.05, indicating there were no statistically significant differences. For example, the number of Preventive-behaviour enterprises was not significantly changed after the administrative reform (Table 4). However, the number of Defensive-behaviour enterprises after vertical administrative reform was statistically different from before the reform.

Table 4.

Analysis of variance (ANOVA) test results for Experiment S_A1.

Scenarios^a	Levene’s test statistic	ANOVA
Defensive-behaviour enterprises (S_B1) versus Defensive-behaviour enterprises (S_A1)	1.52 (p = 0.21)	F = 5.71 (p = 0.01)
Preventive-behaviour enterprises (S_B1) versus Preventive-behaviour enterprises (S_A1)	1.51 (p = 0.23)	F = 2.72 (p = 0.10)
Enthusiastic-behaviour enterprises (S_B1) versus Defensive-behaviour enterprises (S_A1)	1.54 (p = 0.22)	F = 2.71 (p = 0.11)

^aS_B1: before vertical reform; S_A1: after vertical reform.

The simulations found that vertical administrative reform alone would marginally motivate enterprises to improve their environmental behaviour, implying that the vertical reform would strengthen environmental regulatory pressures on these industries. Governmental regulations, such as inspections and fines, have long influenced enterprises’ environmental behaviour. The simulation results also indicated that, with increased regulatory pressures, the number of Defensive-behaviour enterprises would decrease. These enterprises are historically reluctant to accept new environmental regulations and they have even tried to oppose them (Liu, 2009). These enterprises apparently believed that new regulations might increase their costs. However, governmental regulations are compulsory, so the enterprises must take pre-emptive environmental management steps.

Thus, the effects of vertical administrative reform on motivating improved environmental behaviour would be limited, which was found by previous studies. Stafford (2002) explored the effects of the US Environmental Protection Agency enforcement protocol on facility compliance with hazardous waste requirements and found that the violations decreased when the penalties increased, although the decrease was small relative to the recommended increase in penalties. In China, Liu (2009) found that, when basic compliance was not an issue, the regulatory system did not induce enterprises to improve their environmental behaviour. Those findings are supported by the within simulation results finding that the numbers of Preventive-behaviour and Enthusiastic-behaviour enterprises would slightly increase after vertical administrative reform.

Numbers of Preventive-behaviour and Enthusiastic-behaviour enterprises would increase if vertical administrative reform were combined with other factors

In Experiment S_A2, the value of Municipal-county-favouritism was randomly set to 2, 3, and 4, and the values of Environmental-assessment-preparedness (2, 3) and Standard-environmental-regulation (1) were kept at the high level of Experiment S_A1. Other factors of influence were changed to the high level, such as Financial-capacity, Yearly-financial-aid-available, and Environmental-awareness-manager. The comparative results found that, during the simulated period, the number of Preventive-behaviour and Enthusiastic-behaviour enterprises apparently would increase, whereas the number of Defensive-behaviour enterprises would decrease (Figure 7). In Experiment S_A2, all of the ANOVA F-statistics’ p-values were <0.05, indicating statistically significant differences between S_A1 and S_A2 (Table 5).

Figure 7.

Simulated numbers of enterprises by type after vertical administrative reform compared to before vertical administrative reform without (a) and with (b) changes to predictive factors.

Table 5.

Analysis of variance (ANOVA) test results for Experiment S_A2.

Scenarios^a	Levene’s test statistic	ANOVA
Defensive-behaviour enterprises (S_A1)versus Defensive-behaviour enterprises (S_A2)	1.55 (p = 0.24)	F = 5.72 (p = 0.01)
Preventive-behaviour enterprises (S_A1)versus Preventive-behaviour enterprises (S_A2)	1.53 (p = 0.22)	F = 5.73 (p = 0.01)
Preventive-behaviour enterprises (S_A1)versus Preventive-behaviour enterprises (S_A2)	1.52 (p = 0.21)	F = 5.72 (p = 0.01)

^aS_A1: after vertical reform; S_A2: after vertical reform and other variables changed.

The simulation results indicated that, after vertical administrative reform, the improvements to enterprises’ financial capacities and managers’ environmental awareness would significantly influence the enterprises’ environmental behaviour. The numbers of Preventive-behaviour and Enthusiastic-behaviour enterprises would increase and the number of Defensive-behaviour enterprises would decrease. More of the enterprises would commit to improving their environmental behaviour, which should be supported with sufficient financial resources. In particular, enterprises might be uncertain about the extent of the market rewards for their environmental protection efforts (e.g. ‘green’ or low-carbon production processes). However, there was no way of knowing how much consumers might be willing to pay for environmentally cleaner products. Thus, financial capacities and subsidies would be important for promoting the success of vertical administrative reform. On the other hand, the extent of the managers’ environmental awareness would have a stronger influence on what was or was not done than formal rules (Peters, 1999), and influential behavioural factors would be interrelated with cognitive elements (Slovic et al., 2007), which influence enterprises’ environmental behaviour (Papagiannakis and Lioukas, 2012). The simulation results support a previous study, and managers’ characteristics might influence decision-making by contributing to biased forecasts, particularly where decision-making power is highly concentrated (Liang and Reiner, 2008).

This study’s simulation results confirm the study’s theoretical and conceptual models. Vertical administrative integration would create the advantages indicated in the experimental results. The number of Defensive-behaviour enterprises would decrease, but the effects would not be large. Vertical administrative reform might not motivate enterprises to improve their environmental behaviour when basic compliance is not a problem for them. Because environmental behaviour is complex, particularly regarding interactions with other agents, enterprises’ internal motivations are the most important factors. In the way that the complexity of the multi-agent model indicated the dynamic, non-linear, and fitness of self-organizing agents, policy-makers might increase insight into the results of changes made by structural administrative reform.

Conclusions and recommendations

This study simulated the relationship between vertical administrative integration of environmental monitoring at the provincial level and industrial enterprises’ environmental behaviour. Employing an agent-based model and using the case of Fujian Province, the simulations found that the number of enterprises with defensive environmental behaviour would decrease after the reform. However, regarding enterprises with preventive or enthusiastic environmental behaviour, there were no statistically significant differences between their behaviour before and after the reform. Therefore, the effects of the reform would be limited. However, the results of a simulation scenario that included changes to managers’ environmental awareness, enterprises’ financial capacities, and governmental subsidies found that the number of Defensive-behaviour enterprises would decrease after reform and that the numbers of Preventive-behaviour and Enthusiastic-behaviour enterprises would increase. Therefore, vertical administrative reform might effectively motivate improved environmental behaviour when other conditions, such as those related to finances or managers’ environmental awareness, are adjusted.

This study’s results provide valuable information for China’s vertical administrative reform efforts. Vertical reform is considered necessary, and it might decrease industries’ defensive environmental behaviour through strict monitoring and evaluation standards. However, the environmental administrative system is a complex multi-agent system and a simple focus on administrative restructuring is not enough to encourage positive changes in environmental behaviour. Agents’ behaviour or characteristics at the industry level or individual level, such as individual enterprises’ internal characteristics, are important influences on environmental behaviour. Therefore, combining administrative restructuring with regulation of those factors might increase the reform’s success. For example, the Chinese government might consider establishing financial policies, such as combining financing instruments, which would benefit enterprises. The simulation results indicated that the managers’ environmental awareness would be significant for promoting preventive or enthusiastic environmental behaviour, and appropriate incentives to support managers’ active participation might strengthen this relationship. A sophisticated policy approach might ease behavioural changes, particularly through institutional incentives that encourage environmental behaviour via business opportunities intended to stimulate long-term environmentally friendly approaches. Lastly, improved governmental regulations are important, particularly those that concern stable, explicit, and executable environmental standards, although strict regulations and fines are still somewhat effective.

Vertical administrative reform requires sufficient systematic implementation because it cannot operate in isolation from the rest of the overall system; it is a part of the complex and interactive economic, social, and environmental societal system. Thus, other changes also require attention and support, including the broad implementation of an environmental culture and behaviour, such as consumers’ willingness to pay for ‘green’ products, which requires the engagement of the educational system and improvement of the public’s impression of environmental policy.

This study has some limitations, which the author plans to address in future research. First, this study did not include a possible lag-effect or the exhaustive group of factors that likely influence industrial environmental behaviour. Second, actual agents are more intelligent than simulated agents, and the factors influencing their actions are more complex than the model could include. Future studies should also include relatively more reflexive institutions emerging from agents’ behaviour. Nevertheless, this exploratory study provides a foundation for further investigation, and its results will help advance our understanding of vertical administrative reform and pollution-related environmental management in China.

Footnotes

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

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Yong Liu is a Professor at the Sichuan University of China, School of Economics. His research interests include public environmental policy and low carbon economics, local government, and collaborative governance.

References

Ajzen

(2001) Perceived behavioral control, self-efficacy, locus of control, and the theory of planned behavior. Journal of Applied Social Psychology 32(4): 665–683.

Andersson

Aronsson

Wikström

(2004) Testing for vertical fiscal externalities. International Tax and Public Finance 11(3): 243–263.

Balci

(1998) Handbook of Simulation: Principles, Methodology, Advances, Applications, and Practice, Chapter Verification, Validation and Testing. New York: John Wiley and Sons.

Bovaird

(2008) Emergent strategic management and planning mechanisms in complex adaptive systems. Public Management Review 10(3): 319–340.

Bowman

Ambrosini

(1997) Perceptions of strategic priorities, consensus and firm performance. Journal of Management Studies 34(2): 241–258.

Carnovale

Yeniyurt

Rogers

(2017) Network connectedness in vertical and horizontal manufacturing joint venture formations: A power perspective. Journal of Purchasing and Supply Management 23(2): 67–81.

Caswell

Roberts

Lin

(1994) Opportunities to market food safety. In Schertz

Daft

(eds) Food and Agricultural Markets: The Quiet Revolution. Washington, DC: National Planning Association, pp. 12–17.

Chen

(2016) Orderly and effectively promote the reform of local environmental protection management system. Environmental Protection 18: 3–6.

China Statistical Yearbook (2016) Beijing: China Statistics Press.

10.

Christensen

Lægreid

(2008) The challenge of coordination in central government organizations: The Norwegian case. Public Organization Review 8(2): 97–116.

11.

Costa-Font

Rico

(2006) Vertical competition in the Spanish national health system. Public Choice 128(3): 477–498.

12.

Eppel

(2012) What does it take to make surprises less surprising? Public Management Review 14(7): 881–902.

13.

Eppel

(2017) Complexity thinking in public administration’s theories-in-use. Public Management Review 19(6): 845–861.

14.

Fujian Statistical Yearbook (1999) Beijing: China Statistics Press.

15.

Fujian Statistical Yearbook (2012) Beijing: China Statistics Press.

16.

Grauberger

Kimms

(2015) Revenue management under horizontal and vertical competition within airline alliances. Omega 59(2): 228–237.

17.

Guan

Huang

Liu

Chen

(2005) Environmental behavior and analysis of driving model for printing and dyeing enterprises in Taihu Basin. Journal of Lake Science 17(4): 351–355.

18.

Haveri

(2006) Complexity in local government change. Public Management Review 8(1): 31–46.

19.

Hennessy

(1996) Information asymmetry as a reason for food industry vertical integration. American Journal of Agricultural Economics 78: 1034–1043.

20.

Hopkins

(2009) Challenges for the integration of mass drug administrations against multiple ‘neglected tropical diseases’. Annals of Tropical Medicine and Parasitology 103(S1): 23–31.

21.

Jabes

Jans

Frazer-Jans

Zussman

(1992) Managing in the Canadian and Australian public sectors: A comparative study of the vertical solitude. International Review of Administrative Sciences 58(1): 5–21.

22.

Jiao

Alon

Cui

(2011) Environmental dynamism, innovation, and dynamic capabilities: The case of China. Journal of Enterprising Communities People and Places in the Global Economy 5(2): 131–144.

23.

Kawonga

Blaauw

Fonn

(2012) Aligning vertical interventions to health systems: A case study of the HIV monitoring and evaluation system in South Africa. Health Research Policy and Systems 10(1): 1–13.

24.

Kawonga

Fonn

Blaauw

(2013) Administrative integration of vertical HIV monitoring and evaluation into health systems: A case study from South Africa. Global Health Action 6(2): 157–168.

25.

Kilmer

Andre

Tjiii

(2001) Pesticide residues and vertical integration in Florida strawberries and tomatoes. Agribusiness 17(2): 213–226.

26.

Lane

Wallis

(2009) Strategic management and public leadership. Public Management Review 11(1): 101–120.

27.

Law

Kelton

(1991) Simulation Modeling and Analysis. New York: McGraw-Hill.

28.

(2011) Analysis of vertical management models in environmental enforcement. Industrial Safety and Environmental Protection 37(3): 19–20.

29.

(2016) China restructures its environmental agencies from the bottom up. Sixth Tone 28 April. http://www.sixthtone.com/news/786/china-restructures-its-environmental-agencies-bottom

30.

Liang

Reiner

(2008) Behavioral issues in financing low carbon power plants. Energy Procedia 1: 4495–4502.

31.

Liddle

(2018) Aligning vertical structures and horizontal relationships: Collaborative leadership and accountability mechanisms to enhance economic growth in England. International Review of Administrative Sciences 84(4): 659–674.

32.

Liu

(2009) Investigating external environmental pressure on firms and their behavior in Yangtze River Delta of China. Journal of Cleaner Production 17(16): 1280–1286.

33.

Liu

(2017) Industrial pollution resulting in mass incidents: Urban residents’ behavior and conflict mitigation. Journal of Cleaner Production 166(10): 1253–1264.

34.

Liu

Cui

(2011) Study on the relationship between governmental subsidy and corporate recycle ability: A system dynamics approach. Ecological Economy 6: 92–95.

35.

Liu

(2012) The dynamic study on firm’s environmental behavior and influencing factors: An adaptive agent-based modeling approach. Journal of Cleaner Production 37: 278–287.

36.

Loertscher

Reisinger

(2014) Market structure and the competitive effects of vertical integration. RAND Journal of Economics 45(3): 471–494.

37.

Rui

(2015) A springboard into politics: Do Chinese entrepreneurs benefit from joining the government-controlled business associations? China Economic Review 36: 166–183.

38.

Medema

(1992) Transactions, transaction costs, and vertical integration: A re-examination. Review of Political Economy 4(3): 291–316.

39.

Mick

Morlock

Salkever

Malitz

Wise

Jones

(1993) Horizontal and vertical integration-diversification in rural hospitals: A national study of strategic activity, 1983–1988. Journal of Rural Health 9(2): 99–119.

40.

Moore

(1995) Creating Public Value: Strategic Management in Government. Cambridge, MA: Harvard University Press.

41.

Murillo-Luna

Garcés-Ayerbe

Rivera-Torres

(2011) Barriers to the adoption of proactive environmental strategies. Journal of Cleaner Production 19(13): 1417–1425.

42.

Mussa

Pfeiffer

Gloyd

Sherr

(2013) Vertical funding, non-governmental organizations, and health system strengthening: Perspectives of public sector health workers in Mozambique. Human Resources for Health 11(1): 11–26.

43.

Niestroy

(2005) Sustaining Sustainability: A Benchmark Study on National Strategies towards Sustainable Development and the Impact of Councils in Nine EU Member States. Utrecht: Lemma.

44.

Nunan

Campbell

Foster

(2012) Environmental mainstreaming: The organizational challenges of policy integration. Public Administration and Development 32(3): 262–277.

45.

Papagiannakis

Lioukas

(2012) Values, attitudes and perceptions of managers as predictors of corporate environmental responsiveness. Journal of Environmental Management 100(15): 41–51.

46.

Peters

(1999) Institutional Theory in Political Science: The New Institutionalism. New York: Pinter.

47.

Reijnders

(2003) Policies influencing cleaner production: The role of prices and regulation. Journal of Cleaner Production 11(3): 333–338.

48.

Rhodes

(2008) Complexity and emergence in public management. Public Management Review 10(3): 361–379.

49.

Robert

Gunningham

Thornton

(2003) Explaining corporate environmental performance: How does regulation matter? Law and Society Review 37(1): 51–90.

50.

Slovic

Finucane

Peters

MacGregor

(2007) The affect heuristic. European Journal of Operational Research 177: 1333–1352.

51.

Song

Wang

Cen

(2015) Comprehensive efficiency evaluation of coal enterprises from production and pollution treatment process. Journal of Cleaner Production 104: 374–379.

52.

Stafford

(2002) The effect of punishment on firm compliance with hazardous waste regulations. Journal of Environmental Economics and Management 44(2): 290–308.

53.

Sudarmadi

Suzuki

Kawada

Netti

Soemantri

Tugaswati

(2001) A survey of perception, knowledge, awareness, and attitude in regard to environmental problems in a sample of two different social groups in Jakarta, Indonesia. Environment, Development and Sustainability 3: 168–183.

54.

Sumerall

(1995) Vertical integration in medical settings: A brief introduction to its potential effects on professional psychology. Journal of Clinical Psychology in Medical Settings 2(4): 399–406.

55.

Teisman

Klijn

E-H

(2008) Complexity theory and public management. Public Management Review 10(3): 287–297.

56.

Uddin

Mori

Adhikari

(2019) Participatory budgeting in a local government in a vertical society: A Japanese story. International Review of Administrative Sciences 85(3): 490–505

57.

Unger

Green

(2003) A code of best practice for disease control programmes to avoid damaging health care services in developing countries. International Journal of Health Planning and Management 18(S1): S27–S39.

58.

Vita

(2000) Regulatory restrictions on vertical integration and control: The competitive impact of gasoline divorcement policies. Journal of Regulatory Economics 18(3): 217–233.

59.

Wan

(2017) The negative impact of product variety: Forecast bias, inventory levels, and the role of vertical integration. International Journal of Production Economics 186: 123–131.

60.

Wang

Wan

Clement

Begun

(2001) Managed care, vertical integration strategies and hospital performance. Health Care Management Science 4(3): 181–191.

61.

Wescott

(2009) Stimulating vertical integration in coastal management in a federated nation: The case of Australian coastal policy reform. Coastal Management 37(6): 501–513.

62.

World Health Organization (2008) Health Metrics Network: Frameworks and Standards for Country Health Information Systems. Geneva: World Health Organization.