Free Access

Policy Uncertainties, Water Technology and Economic Growth

Fan Yang

https://orcid.org/0000-0002-9217-7916

School of Economics and Management, Southeast University, Nanjing 211189, P. R. China

E-mail Address: 101011572@seu.edu.cn

Search for more papers by this author

Jingyi Ge

School of Arts, Humanities and Social Science, University of Melbourne, Melbourne 3000, Australia

E-mail Address: jgge1@student.unimelb.edu.au

Search for more papers by this author

, and

Xiaomin Wang

School of Business and Economics, The University of Hong Kong, Pokfulan Road, Hong Kong, Hong Kong S.A.R.

E-mail Address: xiaomin.wang@hku.hk

Search for more papers by this author

https://doi.org/10.1142/S2382624X23400076Cited by:1 (Source: Crossref)

This article is part of the issue:

Special Issue on Economics of Water Technology Diffusion and Policy Uncertainties
Guest Editor: Zhanhong Wan

Abstract

Environmental economists have long been interested in the relationship between policy enforcement, technological progress, and economic growth. A multidirectional process exists where factors such as the economy, policy, technology, and environment interactively work. Based on the balanced panel data of 261 prefecture-level cities in China from 2000 to 2020, this study innovatively examines the diversified channels in the policy–technology system within the structural equation modeling framework. Conclusions can be drawn as follows. First, the policy uncertainties curb the technological improvement in the water sector, particularly in southern Chinese cities. Second, the incubation and adoption of water technology are closely intertwined, significantly affecting water quality. By clarifying the interaction of the factors in water technology diffusion, this study provides a comprehensive framework to investigate China’s water economics and policy uncertainties.

Keywords:

JEL: C30, Q53, Q58, R11

1. Introduction

The quantity and quality of water resources are closely related to China’s national economy and people’s standards of living (Zhang et al.2022; Zhu et al.2015). Water shortage, water pollution, and the deterioration of the ecological environment of water have long been restrictive factors for economic development and public health (Lu et al.2017; Wei et al.2022). With the unbalanced development of China’s economy, the goals and methods of water conservation and pollution control in different regions constantly change (Doerfliger et al.1999; Hamouda 2006; Nesta et al.2014; Palmquist 1993; Sun et al.2021).

From the macro perspective, China began to explore water resource planning and allocation utilization via various policies in 1984. In the 1980s and 1990s, policies, regulations, and standards targeting surface water and groundwater were successively formulated and implemented, centering on the Water Pollution Prevention and Control Law of the People’s Republic of China (Li 2022). In 2021, the National Development and Reform Commission and the Ministry of Water Resources jointly issued the Plan for Constructing a Water-Saving Society, which formulated new policies and regulations related to water (Xie and Lu 2022).

From a micro perspective, water technology incubation, diffusion, and adoption are frequently guided by water policies (Noailly 2012; Rustico and Dimitrov 2022). The effect of policy guidance, including production, allocation, and pollution, has expanded as water resources have become scarcer in quantity and quality (Wiser and Pickle 1998). While many scholars have explored policy enforcement in the water sector, the effect of policy uncertainties has been largely ignored until recently. This omission has occurred even though business decision-makers cannot estimate the risks and opportunities of new water technologies without consistency in government policy (Yuan Yuan 2022).

Considering the linkage of factors such as policy, technology, and economy, the relationship between policy uncertainties and innovation in the water sector shows a complicated relationship (Jayasiri et al.2022). The variables in the system influence each other, leading to many different views in academia.

In understanding how policy affects water technology development, deconstructing a complex system in which policy, technology, environmental quality, economics, and resources interact is critical. Accurately describing the fundamental interrelated elements and their interactions is vital to effectively characterize the relationship between core variables (Jiang Zhongyi 2006). Water technology development is in a system including the following interconnectivities. First, policy uncertainty in the water sector is an endogenous variable, determined by the necessity and possibility to change the policy and is confined by the region’s economic development (Verschuur et al.2021). Second, water-related innovation is highly sensitive to the government’s policies since the business decision-maker must estimate the profits, costs, and risks before investing in the innovation (Hamouda et al.2009; Hamouda 2006; Wei et al.2022). Third, the incubation and adoption of new water technology are different steps in water innovation, yet they are intertwined (Beal and Bohlen 1957; Grimaldi and Grandi 2005). Fourth, while population inflow burdens the water environment, water-related technology helps relieve the adverse effects (Al-Kaabi et al.2021; Zhai et al.2021). Fifth, the regional economy is supported by the water resources and population size and confined by the ecological carrying capacity of the water system (Deng et al.2021; Wang et al.2021). Sixth, population is endogenous and depends on many environmental, social and economic factors (Cheng and Duan 2021). In our system, economic development, water quality and water resource abundance are possible influencing factors of the population (Rahman and Alam 2021). Figure 1 presents the overall influence mechanism in a system, including water policy, water technology, and socio-economic factors.

Figure 1. The Influence Mechanism of Water Policy and Water Technology

In summary, the research on water technology diffusion and policy uncertainties should be placed in a multidimensional system with many complex interlocking factors. Most existing studies simplify this multidimensional relationship between technological development and policy consistency into a two-dimensional linear equation and only perform regression analysis on the water technology and water policy, taking factors such as technology, resource endowment, and population as the control variables. The conclusions drawn in this manner mix the direct and indirect effects of multiple factors and the two-way causal relationship. Furthermore, ignoring the endogeneity leads to bias.

This study addresses the following research questions. How to construct a framework for studying a system containing water policy uncertainties, water technology incubation, and adoption? How to analyze the functioning of the interacting and interrelated elements? What is the relationship between policy uncertainties and water technology? How do water technology incubation and adoption link to each other?

The structural equation model (SEM) is a widely used estimation for analyzing complex systems’ influence mechanisms. From the data perspective, the SEM can properly handle problems such as non-normality, non-independence, and heteroscedasticity. From the estimation perspective, the SEM adopts the full information maximum likelihood methods to eliminate the bias resulting from limited information. Since policy uncertainties are measured as dummy variables, we use the generalized SEM to handle the endogenous problem. This study’s marginal contribution is to comprehensively explore the multidimensional influence mechanism of water policy uncertainties, water technology incubation, and technology adoption in a socio-economic system, providing decision-making reference for maximizing the utility of China’s water policy.

The remainder of this study is structured as follows. The second part is the literature review; the third part is the research design and data description; and the fourth part is the generalized structural equation model (GSEM) analysis, including standardization analysis, goodness-of-fit test, reliability, results, and discussion. Finally, the fifth part presents this study’s conclusions and implications.

2. Literature Review

2.1. Policy and technology diffusion

The impact of policy on technology has attracted widespread attention in economics and sociology (Caragliu and Del Bo 2019; Noailly 2012; Rustico and Dimitrov 2022). As a fast-growing field, technology policy theory provides detailed explanations for policy’s role in supporting, enhancing, and developing technology (Branscomb 1997; Dodgson and Bessant 1997). The impact of policy on technology diffusion can be seen in three aspects. First, policies are essential for building the confidence of startups, which are essential in innovation incubation, diffusion, and adoption (Huang et al.2022). Second, policies change the budget constraints of entrepreneurs since the technological projects supported by the policies and regulations can obtain subsidies from the government much easier than its competitors (Wiser and Pickle 1998). Third, policies change the direction of technological innovation; using reward and punishment methods, governments at all levels effectively reshape the focus of technological enterprises (Kyaw 2022).

2.2. Water technology

Despite the evidence that policy changes innovation and technology, relatively little is known about the uniqueness of water resources due to their vulnerability, irreplaceability, and prevalence (Chen and Li 2022). The literature in hydrogeology shows that water resources are unique due to their fragility (Doerfliger et al.1999; Palmquist 1993; Thompson Jr. 1993). In China, the problem of water vulnerability is urgent (Cheng et al.2022). First, the primary threat to China’s water resources is global warming, which alters the spatial and temporal distribution pattern of water resources (Change 2001). Second, China’s water resources are disturbed by changes in human activities resulting from emerging consumerism and the continuously reshaping social structure (Kahrl et al.2005). Third, the uneven distribution of China’s water resources has intensified the contradiction between water supply and demand (Tian et al.2022). Some researchers have developed a systematic scheme to characterize the importance of water from its irreplaceability and prevalence (Haile et al.2022). As a vital part of people’s daily lives and considering China’s huge population, water resources are not easy to be substituted (Wang et al.2022). As an essential production factor, the irreplaceability of China’s water resources is even more prominent, given the country’s role as the world’s factory (Yang et al.2022). Due to the inherent vulnerability and importance of China’s water resources, policymakers need to reduce external shock and disturbances (Yuan Yuan 2022), thus encouraging new technology to improve water quality and quantity and prevent water hazards (Thompson Jr. 1993).

2.3. Policy uncertainties

Despite the evidence in the literature that policy enforcement significantly impacts technology, the effects of policy uncertainties are yet to be explored. Studies from various disciplines have indicated the differences between policy enforcement and policy uncertainties since the latter describes the fluctuation rather than the level of the former (Kyaw 2022; Ye et al. 2022; Zhang 2020). Policy uncertainty is the instability in the timing, content, and potential impact of policy decisions (Kyaw 2022). Furthermore, policy uncertainties change the whole technology process, including incubation and adoption (Kanger et al.2019).

From the perspective of technology incubation, Grimaldi and Grandi (2005) divided business incubations into several steps, including a business plan, elaboration, access to finance, hosting, and training. We believe that policy uncertainties can affect every step of incubation. In the stage of plan and elaboration, considering the costs of the changing ideas and designs, a policy with good intentions but lacking stability can distort technology elaboration (Huang et al.2022). When startups try to access finance, uncertainties in policies and regulations discourage firms from investing in new ideas by eliminating people’s confidence (Yuan et al.2022). In hosting and training, a constantly changing policy always leads to extra friction costs, creating difficulty for the company to expand the coverage of new technology (Li and Li 2022).

Focusing on technology adoption, Beal and Bohlen (1957) proposed the concept of the technology adoption life circle, which they classified into four stages: innovation, early adoption, majority, and non-adoption. In the early adoption period, policy uncertainties narrow people’s vision by preventing them from noticing how this technology could be applied to solve problems (Khanam and Daim 2021). In the majority period, policy uncertainties make the practical application and market acceptance difficult, leading to less technology diffusion in the private sector (Kyaw 2022).

2.4. Policy uncertainties and water technology

Battamo et al. (2022) highlighted the importance of well-designed policies to improve water adaption capacity and decrease its sensitivity and vulnerability; however, their analyses have focused on policy enforcement rather than policy uncertainties. Hamouda et al. (2009) emphasized the vulnerability of water facing institutional uncertainties. They divided water resource vulnerability into five aspects: hydro-physical and ecological vulnerability, water infrastructure and management vulnerability, geopolitical and institutional vulnerability, economic and sociocultural vulnerability, and population vulnerability, in which instability from the institution was deemed to be a dominant source (Hamouda et al.2009; Hamouda 2006).

In conclusion, the relationship between water technology and policy uncertainties remains unexplored. This study examines the connection between China’s policy uncertainties, water technology incubation, and water technology adoption. Wei et al. (2022) proposed an index for water resource vulnerability assessment through a structural lens, defining the vulnerability as a combination of water system exposure, sensitivity, and adaptability. We also notice that policy uncertainties and water technology diffusion are indirectly connected via some intermediaries and interact in feedback loops (Kanger et al.2019). First, economic development simultaneously reshapes water technology and policy (Karakaya et al.2014). Second, adopting new water technology can help improve water quality, which is essential in designing water policy (Palm 2022). Third, the economy and environment are a system in which economic, environmental, and technological resources and other factors are intertwined (Yang et al.2022). Therefore, we conduct our research in a socio-economic framework, performing systematic comparative analyses on the relationship between water technology and policy uncertainties.

3. Research Design and Data Description

3.1. Framework of structural equation modeling (SEM)

This study uses SEM to describe the multidimensional interaction channels between various components of the environmental and economic system. The measurement model described below can be used to express the framework of the structural equation :

Y = Λ y η + ε, <math display="block" altimg="eq-00001.gif"><mi>Y</mi><mo>=</mo><mi mathvariant="normal">Λ</mi><mi>y</mi><mi>η</mi><mo>+</mo><mi>ε</mi><mo>,</mo></math> (1)

X = Λ x ξ + δ, <math display="block" altimg="eq-00002.gif"><mi>X</mi><mo>=</mo><mi mathvariant="normal">Λ</mi><mi>x</mi><mi>ξ</mi><mo>+</mo><mi>δ</mi><mo>,</mo></math> (2)

η = β η + Γ ξ + ζ . <math display="block" altimg="eq-00003.gif"><mi>η</mi><mo>=</mo><mi>β</mi><mi>η</mi><mo>+</mo><mi mathvariant="normal">Γ</mi><mi>ξ</mi><mo>+</mo><mi>ζ</mi><mo>.</mo></math> (3)

Structural equations may deal with latent variables — which cannot be directly measured but can be inferred using mathematical models — and explicit variables, which can be precisely stated and quantified to obtain reliable measurements of logical relationships (Gunzler et al.2013). The relationship between the latent and explicit variable series is described by Eqs. (1) and (2). Among these, X and Y are the explicit exogenous variables. Furthermore, $η$ and $ξ$ are endogenous and exogenous latent variables, respectively, and $δ$ and $ε$ represent the error terms.

The correlation between endogenous and exogenous latent variables is shown in Eq. (3). In Eq. (3), the $β$ coefficient matrix represents the interaction between various endogenous and latent variables, while the $Γ$ coefficient matrix shows the impact of exogenous latent variables on the endogenous ones. Moreover, $ζ$ represents the matrix of error terms. Similar to ordinary least squares regression, the SEM assumes that the disturbance term and the explanatory variable are independent. SEM further assumes the independence between the residuals of equations and the independent variables (Wang et al.2016).

The main benefit of SEM is the processing of multiple equations. Real-world policy uncertainty and the spread of water technologies is a complex system that requires interdisciplinary explanations. Structural equations can be employed to express the system’s multidimensional interacting connections to address the issue of linear equations’ pseudo-regression.

3.2. Hypotheses

The variables’ interaction described in our introduction and literature review can be interpreted in the following hypotheses. Seven variables are included in Figure 1, in which water resource abundance is the only exogenous variable. The influencing factors of the remaining six endogenous variables are as follows:

	Hypothesis 1: Policy uncertainties are determined by economic development, water quality, and the abundance of water resources (Verschuur et al.2021).
	Hypothesis 2: The abundance of water resources, policy uncertainties, economic development, and the adoption of water technology are all critical factors affecting water technology incubation (Hamouda et al.2009; Hamouda 2006; Wei et al.2022).
	Hypothesis 3: Water technology adoption is jointly determined by water resource abundance, policy uncertainties, economic development, and technology incubation (Beal and Bohlen 1957; Grimaldi and Grandi 2005).
	Hypothesis 4: Water quality is affected by the incubation and adoption of water technology, water resource abundance, and population (Al-Kaabi et al.2021; Zhai et al.2021).
	Hypothesis 5: Economic development is jointly influenced by population, water resources, and environmental factors (Deng et al.2021; Wang et al.2021).
	Hypothesis 6: Urban population is affected by water quality, water resource abundance, and economic development (Cheng and Duan 2021).

3.3. Variables and proxies

This study adopts the balanced panel data of 261 prefecture-level cities in China from 2000 to 2020. The existing literature on policy uncertainty mainly uses the model of Baker et al. (2016) to calculate the value of policy uncertainty. By counting the frequency of relevant keywords appearing in newspapers, they calculate the index of policy uncertainties. However, there are two reasons that we do not follow that approach. One reason is that the keywords concerning “water policy uncertainties” can be rarely found in the news. Though the words related to “water policy” can easily be captured through various qualitative research tools, it’s not easy to define “uncertainties” from these words. In addition, newspaper reports may have biases, especially related to policy instability or policy uncertainty.

In contrast, we measure water policy uncertainty by firstly quantifying the level of water policy, and then calculating the water policy uncertainties using the pattern of mean absolute deviation. First, we specify the enforcement of the water policy. We collect information about the water policy in China’s Five-Year Plan. For China’s prefecture-level cities, the vital water policy is embodied in the country’s Five-Year Plan, which is the country’s core plan. Specifically, the central government selected dozens of cities as controlling unit in its Five-Year Plan. On the one hand, the pollution index such as chemical oxygen demand (COD), total nitrogen, and total phosphorus of controlling units is under close watch by the central government. On the other hand, the controlling units got extra financial support from the government in all administrative ranks. As a result, we choose the dummy variable of whether a city is selected as the controlling unit as a proxy for the water policy enforcement. Second, we measure the uncertainties of water policy. We follow the pattern in constructing the statistics of mean absolute deviation, in which the absolute deviations are aggregated to measure variance and fluctuation of the data (Ardia et al.2021; Białkowski et al.2022). It is worth noting that the fluctuations lead to uncertainties during the process of formulating and implementing policies (Bloom 2014). As our observing period spans from China’s 9th Five-Year Plan (1996–2000) to 13th Five-Year Plan (2016–2020), we can observe clear modifications of water policy in these periods. If the water policy of a region has changed, there will be differences between the policy of the current period and its one-lagged period. Therefore, we use the absolute difference between the value of water policy enforcement ( $x_{i j}$ ) and its one-period-lagged value ( $x_{i, j - 1}$ ) as a proxy for water policy uncertainties.

Both technology incubation and adoption of innovation characterize water technology. We use the number of water technology patents to measure the technology incubation. Not all technological incubations are patented, yet the patented innovation is a prominent part of incubation (Acs and Audretsch 1989). So that the connection between patents and innovations of all categories are positive, significant and stable (Ye 2022). Due to the relevance and reliability of the connection in empirical research, patents are widely used in literature to characterize technological incubation (Feldman and Florida 1994; Gernego et al.2019). Our data is collected from the patent database of China National Knowledge Infrastructure (CNKI), in which we find the information on cities from the institution of applicants and years from the application date. There are invention, utility patent, and design patents in the category of patents. So, we use them as three proxies to measure the water technology incubation.

The concept of water technology adoption has a wide range of denotations. From the perspective of industrial classification, water technology adoption covers technologies used in the range of primary, secondary and tertiary industries. From the view of input–output accounting, water technology adoption includes the pumping and treatment of water resources, as well as the treatment and discharge of wastewater (O’Callaghan 2020). Multiple water technologies are not associated with wastewater treatment. However, we still use the ratio of industrial wastewater treatment to discharge as a proxy for water technology adoption for the following reasons. First, wastewater treatment is the most prominent part of water technology in terms of energy consumption and environmental deterioration since China’s emergence as the world’s factory (Qu et al.2019). Second, the ratio of wastewater centralized treated of sewage work is the nearest proxy for water technology adoption which can be available in prefecture-level cities to the best of our knowledge. Third, due to the heterogeneity of different industries, an aggregation of water technology adoption in agricultural irrigation, industrial manufacturing and commerce may lead to bias (Ahmad et al.2021). As a result, the ratio of wastewater treatment to discharge is used to measure water technology adoption in our study.

The abundance of water resources is conducive to the development of water policy, which we measure with the city’s water resources from each province’s Water Resources Bulletin. For the water quality, we find the data from each city’s Environmental Bulletin. In prefecture-level cities, the water quality of their monitored section is annually reported, providing the percentage of the monitored section that meets the requirement of China’s water environmental quality standards. We gather the quality data on levels I, II, and III to measure the water quality. Furthermore, the data representing economic development and population are gathered from the China City Statistical Yearbook. Some early-year data are missing, and we use Lagrange interpolation for imputation. The operational definition and descriptive statistics of each variable are shown in Table 1. The spiderweb model that presents the correlation of manifest variables and latent variables is shown in Figure 2.

**Table 1. Operational Definition and Descriptive Statistics**
Variables	Abbreviation	Measurement	Operational Definition	Unit	Obs	Mean	S.D.	Min	Max
Policy Uncertainties	policy	policy	$\| X_{i j} - X_{i, j - 1} \|$	N/A	5481	0.27	1.381	0	1
Water Technology Incubation		invention	Number of water technology inventions	Item	5481	135.7	2182	35	525
	techin	utility	Number of water technology utility patents	Item	5481	89.72	1876	32	301
		design	Number of water technology design patents	Item	5481	45.90	957.3	17	137
Water Technology Adoption	techad	tech_adopt	Wastewater treatment/discharge	%	5481	75.27	134.7	12	100
Water resource abundance	resources	resources	Amount of water resources	10⁸m³	5481	23.81	103.5	5	160
Water quality		qualityI	Monitored section meets the Standard I	%	5481	67	567.9	26	100
	quality	qualityII	Monitored section meets the Standard II	%	5481	85	783.4	57	100
		qualityIII	Monitored section meets the Standard III	%	5481	92	954.9	76	100
Economic development	gdp	gdp	GDP	10⁴ yuan	5481	1.75e $+$ 07	2.93e+07	340500	3.27e $+$ 08
Population	popu	popu	Population of the city at the end of the year	10000 person	5481	5.872	0.700	2.773	8.124

3.4. Model construction

Following the relationship of the variables in Figure 1 and the corresponding research hypothesis, we describe the links within the framework of the SEM, where water technology incubation and water quality are taken as latent variables. As the policy uncertainties are categorical variables, we use the GSEM to analyze the links among the variables.

In the part that variables affect policy uncertainties, we use zero-inflated Poisson (ZIP) regression for the following reasons. On the one hand, we can observe a large percentage of zero values in the variable of policy uncertainties beyond the proper numbers that could be estimated from a Poisson model. On the other hand, we believe the number of zeros may be inflated because of the unobserved factor. Specifically, a region’s water policy is jointly affected by its necessity and possibility, while the former is mainly unobserved. Some cities do not need to change the water policy reinforcement; however, we only have data on how many cities changed their water policy and not on the necessity thereof. A zero value of policy uncertainties may come from a city that does not have to change its water policy; it may also result from a city that needs to change its policy but lacks financial support. In a standard Poisson model, these two kinds of cities are identically treated, which would lead to bias. We believe distinguishing between these two groups is necessary using the ZIP regression.

In Eq. (6), we use the ZIP regression to model the two types of zeros. First, we model whether a city must change the water policy as a function of its water resource abundance. Subsequently, we assume that the dependent variable, policy uncertainties, is jointly decided by the economic and environmental factors that can change the policy. According to our theoretical framework and research hypotheses, the measurement model is set as the following equation :

policy= α 1 gdp + α 2 quality + e 1, <math display="block" altimg="eq-00017.gif"><mstyle><mtext mathvariant="normal">policy=</mtext></mstyle><msub><mrow><mi>α</mi></mrow><mrow><mn>1</mn></mrow></msub><mstyle><mtext mathvariant="normal">gdp</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>2</mn></mrow></msub><mstyle><mtext mathvariant="normal">quality</mtext></mstyle><mo>+</mo><msub><mrow><mi>e</mi></mrow><mrow><mn>1</mn></mrow></msub><mo>,</mo></math> (4)

techin = α 3 policy + α 4 gdp + α 5 techad + α 6 resources + e 2, <math display="block" altimg="eq-00018.gif"><mstyle><mtext mathvariant="normal">techin</mtext></mstyle><mo>=</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>3</mn></mrow></msub><mstyle><mtext mathvariant="normal">policy</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>4</mn></mrow></msub><mstyle><mtext mathvariant="normal">gdp</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>5</mn></mrow></msub><mstyle><mtext mathvariant="normal">techad</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>6</mn></mrow></msub><mstyle><mtext mathvariant="normal">resources</mtext></mstyle><mo>+</mo><msub><mrow><mi>e</mi></mrow><mrow><mn>2</mn></mrow></msub><mo>,</mo></math> (5)

techad = α 7 policy + α 8 gdp + α 9 techin + α 10 resources + e 3, <math display="block" altimg="eq-00019.gif"><mstyle><mtext mathvariant="normal">techad</mtext></mstyle><mo>=</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>7</mn></mrow></msub><mstyle><mtext mathvariant="normal">policy</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>8</mn></mrow></msub><mstyle><mtext mathvariant="normal">gdp</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>9</mn></mrow></msub><mstyle><mtext mathvariant="normal">techin</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>1</mn><mn>0</mn></mrow></msub><mstyle><mtext mathvariant="normal">resources</mtext></mstyle><mo>+</mo><msub><mrow><mi>e</mi></mrow><mrow><mn>3</mn></mrow></msub><mo>,</mo></math> (6)

quality = α 11 techin + α 12 techad + α 13 popu + α 14 resources + e 4, <math display="block" altimg="eq-00020.gif"><mstyle><mtext mathvariant="normal">quality</mtext></mstyle><mo>=</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>1</mn><mn>1</mn></mrow></msub><mstyle><mtext mathvariant="normal">techin</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>1</mn><mn>2</mn></mrow></msub><mstyle><mtext mathvariant="normal">techad</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>1</mn><mn>3</mn></mrow></msub><mstyle><mtext mathvariant="normal">popu</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>1</mn><mn>4</mn></mrow></msub><mstyle><mtext mathvariant="normal">resources</mtext></mstyle><mo>+</mo><msub><mrow><mi>e</mi></mrow><mrow><mn>4</mn></mrow></msub><mo>,</mo></math> (7)

gdp = α 15 resources + α 16 popu + e 5, <math display="block" altimg="eq-00021.gif"><mstyle><mtext mathvariant="normal">gdp</mtext></mstyle><mo>=</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>1</mn><mn>5</mn></mrow></msub><mstyle><mtext mathvariant="normal">resources</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>1</mn><mn>6</mn></mrow></msub><mstyle><mtext mathvariant="normal">popu</mtext></mstyle><mo>+</mo><msub><mrow><mi>e</mi></mrow><mrow><mn>5</mn></mrow></msub><mo>,</mo></math> (8)

popu = α 17 quality + α 18 gdp + α 19 resources + e 6 . <math display="block" altimg="eq-00022.gif"><mstyle><mtext mathvariant="normal">popu</mtext></mstyle><mo>=</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>1</mn><mn>7</mn></mrow></msub><mstyle><mtext mathvariant="normal">quality</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>1</mn><mn>8</mn></mrow></msub><mstyle><mtext mathvariant="normal">gdp</mtext></mstyle><mo>+</mo><msub><mrow><mi>α</mi></mrow><mrow><mn>1</mn><mn>9</mn></mrow></msub><mstyle><mtext mathvariant="normal">resources</mtext></mstyle><mo>+</mo><msub><mrow><mi>e</mi></mrow><mrow><mn>6</mn></mrow></msub><mo>.</mo></math> (9)

This paper takes the prefecture-level cities as the basic unit of analysis; from the water resources perspective, heterogeneity exists within China’s prefecture-level cities. Northern Chinese cities lack water, while the south is rich in water resources; therefore, the South-to-North Water Diversion Project diverts 44.8 billion cubic meters of water annually to the dry north. The huge difference in water resources leads to their water policy and technology diversion. Therefore, in addition to the overall estimation, we divide the 261 city samples into northern and southern cities and explore the link between water technology diffusion and policy uncertainties in different groups.

4. Structural Equation Analysis

4.1. Standardization

The restriction of the linear regression approach is that the magnitude and the variable unit impact the parameter estimate outcome. We employ the z-score approach to further normalize the regression coefficients, removing this effect and facilitating the interpretation and comparison of the findings. Equation (10) shows the standardized formula, where $X_{i}$ is the original data and $X_{i}^{'}$ is the standardized value.

X′i=Xi−1n∑ni=1Xi√1n∑ni=1(Xi−1n∑ni=1Xi)2.<math display="block" altimg="eq-00025.gif"><msubsup><mrow><mi>X</mi></mrow><mrow><mi>i</mi></mrow><mrow><mi>′</mi></mrow></msubsup><mo>=</mo><mfrac><mrow><msub><mrow><mi>X</mi></mrow><mrow><mi>i</mi></mrow></msub><mo>−</mo><mfrac><mrow><mn>1</mn></mrow><mrow><mi>n</mi></mrow></mfrac><munderover accentunder="true" accent="true"><mrow><mo>∑</mo></mrow><mrow><mi>i</mi><mo>=</mo><mn>1</mn></mrow><mrow><mi>n</mi></mrow></munderover><msub><mrow><mi>X</mi></mrow><mrow><mi>i</mi></mrow></msub></mrow><mrow><msqrt><mrow><mfrac><mrow><mn>1</mn></mrow><mrow><mi>n</mi></mrow></mfrac><munderover accentunder="true" accent="true"><mrow><mo>∑</mo></mrow><mrow><mi>i</mi><mo>=</mo><mn>1</mn></mrow><mrow><mi>n</mi></mrow></munderover><msup><mrow><mo stretchy="false">(</mo><msub><mrow><mi>X</mi></mrow><mrow><mi>i</mi></mrow></msub><mo>−</mo><mfrac><mrow><mn>1</mn></mrow><mrow><mi>n</mi></mrow></mfrac><munderover accentunder="true" accent="true"><mrow><mo>∑</mo></mrow><mrow><mi>i</mi><mo>=</mo><mn>1</mn></mrow><mrow><mi>n</mi></mrow></munderover><msub><mrow><mi>X</mi></mrow><mrow><mi>i</mi></mrow></msub><mo stretchy="false">)</mo></mrow><mrow><mn>2</mn></mrow></msup></mrow></msqrt></mrow></mfrac><mo>.</mo></math>(10)

The independent variable’s effect on the dependent variable, measured in standard deviation units, is represented by the standardized coefficient processed in Eq. (10). The standardized coefficient can be understood as the correlation coefficient between the explanatory and explained variables, where both variables are assessed by a single indicator (Schumacker and Lomax 2004).

4.2. Goodness-of-fit and reliability

The statistics in Table 5 assess the model’s goodness-of-fit from various perspectives. Table 2 shows the goodness-of-fit in the groups of all cities, northern cities, and southern cities in China, indicating that the goodness-of-fit index is generally within acceptable bounds; thus, the empirical analysis supports our research hypothesis.

**Table 2. Goodness-of-Fit**
		Value
Goodness-of-Fit	Statistics	Overall	Northern Cities	Southern Cities
Likelihood ratio	$p > chi 2$	0.000	0.000	0.000
RMSEA	90% CI, lower bound	0.000	0.000	0.000
Information criteria	AIC	38291.361	17543.622	23457.664
	BIC	36759.428	17634.108	23627.821
Baseline comparison	CFI	1.000	1.000	1.000
Size of residuals	CD	0.853	0.847	0.693

The likelihood ratio shows the $X^{2}$ test of the benchmark model versus the saturation model. Root mean square error of approximation (RMSEA) represents the square root of the average approximation error. Akaike information criterion (AIC), Bayesian information criterion (BIC), and comparative fit index (CFI) represent the Akaike information index, Bayesian information index, and comparative fit index, respectively. CD represents the coefficient of determination.

Table 2 shows that the P value of the $X^{2}$ test, which indicates that the GOF of the structural equations of all the cities, northern Chinese cities, and southern Chinese cities are significantly better than those in the benchmark models. The RMSEA index is less than 0.05, showing the significance of the three structural equations.

The AIC and BIC indices are primarily used to compare model goodness-of-fit. In China, northern cities have the best fit, indicating their development trajectories are similar and more homogeneous. Southern cities have a slightly weaker fit, indicating that the endowments of cities in their groups differ, and the AIC and BIC indices of the entire country have the strongest intra-group heterogeneity due to their different city categories. The structural equation comprising the sample of all cities has the lowest accuracy and information, confirming the statistical importance of categorizing cities into two groups.

The CFI index measures the overall relationship between the variables. The comparative fitting index of Table 2 is close to one, suggesting a strong correlation between the variables in the three structural equations. Regarding the residuals expressed by the CD coefficient, the national residuals are similar to those of northern cities, with the biggest residuals shown in southern cities.

Composite reliability (CR) and Cronbach’s alpha were used to examine the reliability of the factors to the constructs. Table 3 shows that most CR and AR are higher than 0.6, indicating that their factors correctly measure the water technology incubation and water quality (Fornell and Larcker 1981).

**Table 3. Reliability**
Construct	Items	Factor Loading	CR	CA
water technology incubation	invention	0.76	0.71	0.65
	utility	0.77
	design	0.69
water quality	qualityI	0.61	0.64	0.67
	qualityII	0.76
	qualityIII	0.51

4.3. Results and discussion

This paper researches the interaction mechanism between water policy uncertainties and water technology incubation, diffusion and adoption. We use STATA 17.0 to analyze the data from 5481 sample points in 261 prefecture-level cities. In this paper, the Zero-inflated Poisson regression and generalized structural equation model have been used to estimate the parameters of the standardized path coefficients. The 1000 repeated bootstrap was used to obtain the standard errors of the estimations. The results of all the cities, northern Chinese cities and southern Chinese cities are shown in Tables 4–6, respectively. In the tables, the factors in the row are the explained variables. Correspondingly, the factors in the column are the explanatory variables. Therefore, those numbers in the tables represent the impact of each column factor on each row factor. Based on the result, most of the standardized coefficients are significant at 10%, which confirms our hypothesis. By comparing all the cities, northern Chinese cities and southern Chinese cities, we found that the regression coefficients are at different levels in these three groups’ values, yet with identical tendencies. That shows the similar mechanism of the three groups.

The first column of Tables 4–6, with policy uncertainties as explained variables, corresponds to the regression equation (4) that uses the validity of Hypothesis 1. The result shows a negative correlation between water quality and policy uncertainties. In other words, in areas with a better environment and higher water quality, the government would more easily achieve a policy consistency. Considering the growth effect, we can identify that the economic growth of these three different groups in the research results in lower policy uncertainties, especially the northern Chinese cities. The water resource abundance, which is regarded as a zero-inflation factor of policy uncertainties to characterize the necessity to maintain a consistent policy, shows a clear impact on the dependent variable. Therefore, it is proved that the policy consistency is not only determined by its possibility but also by its necessity.

**Table 4. Standardized Analysis (All Cities)**
	Policy Uncertainties	Water Technology Incubation	Water Technology Adoption	Water Quality	Economic Development	Population
Policy uncertainties		−0.392***	−0.042***
		(0.0749)	(0.1778)
Water technology incubation			0.171***	0.393*
			(0.5279)	(0.2988)
Water technology adoption		0.139***		0.059***
		(0.7480)		(0.1329)
Water quality	−0.499***				−0.182**	−0.028
	(0.0394)				(0.0463)	(0.8428)
Economic development	−0.109***	0.342**	0.218***			0.368***
	(0.0481)	(0.3706)	(0.0765)			(0.1267)
Population				−0.277***	0.539***
				(0.1988)	(0.0330)
Water resource abundance	−0.732**	0.573	0.349	0.584	0.337*	0.140
	(0.3580)	(0.7489)	(0.2120)	(0.2726)	(0.0259)	(0.0573)

**Table 5. Standardized Analysis (Northern Chinese Cities)**
	Policy Uncertainties	Water Technology Incubation	Water Technology Adoption	Water Quality	Economic Development	Population
Policy uncertainties		−0.065*	−0.046***
		(0.0311)	(0.0493)
Water technology incubation			0.503***	0.827**
			(0.0152)	(0.0218)
Water technology adoption		0.137**		0.627*
		(0.0172)		(0.0134)
Water quality	−0.703***				−0.038**	−0.172*
	(0.0184)				(0.0264)	(0.0374)
Economic development	−0.129***	0.150***	0.234**			0.425***
	(0.0457)	(0.0274)	(0.0319)			(0.0268)
Population				−0.063***	0.482***
				(0.0259)	(0.0259)
Water resource abundance	−0.489*	0.524	−0.458	0.395*	0.372***	0.047
	(0.6468)	(0.0934)	(0.9091)	(0.4236)	(0.0258)	(0.5328)

**Table 6. Standardized Analysis (Southern Chinese Cities)**
	Policy Uncertainties	Water Technology Incubation	Water Technology Adoption	Water Quality	Economic Development	Population
Policy uncertainties		−0.164**	−0.108*
		(0.0137)	(0.2386)
Water technology incubation			0.264***	0.467*
			(0.0221)	(0.1312)
Water technology adoption		0.162***		0.017***
		(0.0874)		(0.3668)
Water quality	−0.327***				−0.271*	−0.138
	(0.0459)				(0.0368)	(0.2562)
Economic development	−0.125***	0.220***	0.671***			0.458***
	(0.3417)	(0.1285)	(0.0209)			(0.2360)
Population				−0.502**	0.638***
				(0.1201)	(0.0472)
Water resource abundance	−0.623**	−1.294	0.174	0.029	0.424*	0.592
	(0.3849)	(0.0295)	(0.0343)	(0.3156)	(0.0571)	(0.6231)

Note: ***, **, and * are significant at 1%, 5%, and 10%, respectively. The standard errors of the coefficients are in brackets.

The second and third columns of Tables 4–6 show the coefficients of structural equations (5) and (6), in which water technology incubation and adoption are explanatory variables. The equations correspond to Hypotheses 2 and 3. The provided coefficients show different impacts from predictors in all cities as well as the north and south groups. First, there is a close relationship between water policy uncertainties and water technology incubation. Specifically, the change of policy uncertainties in one standardized unit is shown to reduce the predicted value of water technology incubation by 0.392 standardized unit in China. If water policy is repeatedly changing in some areas, entrepreneurs may have to give up their plans, such as investment, diffusion, and adoption of new technological innovation, to avoid unknown risks from the water policy. Second, the adoption of new technological innovations is also sensitive to the water policy uncertainties, with its country-wide standardized unit coefficient as $- 0.042$ . Third, we found the mutual effect between technology incubation and adoption, in which the impact from incubation to adoption is relatively higher in terms of standardized units. Fourth, economic growth is found to be conducive to the incubation and adoption water technology, while the effect of water resource abundance is insignificant.

The coefficients of structural equation (7) are shown in the fourth column of Tables 4–6. Corresponding to Hypothesis 4, the predicted outcome of this equation is water quality. First, we found that water quality is changed by technology incubation, adoption and population size. The results of structural equation analysis show that developing water technology incubation, diffusion and adoption could improve water quality. Second, larger population size inevitably results in lower water quality. Third, we found technology incubation and adoption could bring different levels of impact on water quality. The standardized coefficients show that water technology adoption is more effective than water technology incubation in improving water quality. Fourth, the correlation between water quality and water resource abundance is not significant except for northern cities, indicating that the deteriorated water quality can be resulted from human pollution rather than the scarcity of water resources.

The data in the fifth column of the above three tables correspond to Hypothesis 5. They show the coefficients of equation (8) in which economic development is the dependent variable. First, the link between water quality and economic growth caught our attention. The standardized path coefficients of all cities, northern cities and southern cities are $- 0.182$ , $- 0.038$ , and $- 0.271$ , respectively. Therefore, severe water pollution is accompanied by prosperous economic development, which indicates that China’s economic development has not yet reached the turning point of the environmental Kuznets curve. The reason might be that many enterprises prefer to set up factories in areas with preferential policies and relaxed environments rather than areas with favorable and comfortable climates. Among the three groups, the influence of environmental quality on economy is relatively insignificant for southern cities. The standardized path coefficient of southern cities indicates a sign of relative decoupling between economic growth and water pollution, which is different from the other two groups. Second, we found that the population size noticeably contributes to economic growth. This phenomenon is especially obvious in southern Chinese cities. Third, we found that water resource abundance is positively correlated with economic growth, indicating that the resource curse is unapparent in water resources. Instead, the cities’ abundant water resources often facilitate their growth.

The sixth column of Tables 4–6 shows the coefficients of regression equation (9), that are used on the validity of Hypothesis 6. They show how the cities’ population is changed by economic development and water quality. On the one hand, economic development is conducive to the population expansion. When people are viewed as consumers, high-quality and diversified products could attract more people to move into economically developed regions. When people are viewed as labor force, higher income and better employment opportunities are appealing to job-seekers (Li et al.2017; Yunjiang and Xiangdong 2017). On the other hand, we found that the impact of the water quality on population size is insignificant in the groups of all cities and southern cities, while the standardized coefficient of northern cities even shows a negative link. Although people can vote with their feet, they may prefer to stay in urban areas with a good economy rather than a comfortable environment. If higher income can improve marginal utility greatly, people could tolerate the unpleasant environment as much as possible (Yang 2019). Therefore, people prefer to stay in a fast-growing economy and relatively polluted water environment rather than high-water quality areas. In addition, we notice that population is not dependent on water resource abundance. The reason might be that domestic water is not a constraint with the development of water facilities in Chinese cities.

5. Conclusions and Implications

We first outline the conclusion derived from our empirical analysis. Next, we provide the policy implications based on our research results. We then indicate several limitations of our study that can be addressed in future research.

5.1. Conclusions

First of all, one contribution of our study is the quantitative research of China’s water policy uncertainties, which is an innovation to the best of our knowledge. Compared to previous research that focuses on water policy enforcement, our research carries distinctive importance since the effect of policy stability can be different from the effect of policy enforcement. Additionally, we measure water policy uncertainties using China’s city-level data, which allow us to explore the efficiency of policy in a detailed way.

Accumulating studies argued that stochastic policies can work better than deterministic policies (Robins et al.2017; Singh et al.2000; Zhang and Oki 2023; Zhang et al.2009). However, this study helps us to expand the typical viewpoint that insufficient water technology innovation is due to inadequate policy reinforcement. We indicate that inconsistent policies with good intentions could also lead to market failure. Specifically, the instability in timing and content of policy can result in disturbance of water technology, especially in the stage of business plan, elaboration and early adoption.

Another contribution of this research is developing an integrated ecological-economic model that contains the core concepts of policy uncertainties and water technology. In the framework of structural equation modeling, we comprehensively analyzed the endogeneity of water policy and its impact on the whole process of water technology innovation. Besides, we further discover the feedback mechanism from water technology adoption to its incubation, thus confirming a positive feedback loop between the incubation and adoption of water technology.

5.2. Implications

There are three interesting policy implications based on our research results. First, we provide an empirical rationale for the stability and certainty of water policy, which is different from the studies that put emphasis on constantly enhancing water policy. Thus, we conclude that long-term and deterministic policies are more efficient than frequently adjusted incentives. Second, improving the transparency in policy can enhance entrepreneurs’ confidence, which in turn facilitates the incubation and adoption of water technology. Accordingly, water administrative departments should proactively disclose information about water policy changes through government gazettes, official websites, press conferences, newspapers, and other social media. Third, by looking into the policy-technology system, we found a mutual effect between the adoption and incubation of water technology. Water technology adoption is not only supported by water technology incubation in the whole process, but it also creates conditions that inspire the development of water technology incubation, especially in southern cities. Therefore, the feedback mechanism should be considered in the process of water policy design and evaluation.

5.3. Limitations

Our analysis contains several limitations that can be addressed in future research. First, we acknowledge that our proxy for policy uncertainties is evidently simplified. While it is challenging to measure China’s water policy uncertainties, we tried to abstract the policy uncertainties from city-level data which change by the national strategy in these years. The city-level data allowed us to document new facts on water policy uncertainties and explore their impact on water technology. Yet future work could be devoted to expanding the coverage of water policy and investigating the heterogeneous policy effects on province and county levels. Second, the policy-technology system provided here could be regarded as a basic framework of a structural model with economic, environmental and natural resources factors. We leave the improvement of the model to future research.

Acknowledgment

This work was supported by “the Fundamental Research Funds for the Central Universities” (2242022S20033).

References

Ahmad, M, G Jabeen and Y Wu [2021] Heterogeneity of pollution haven/halo hypothesis and environmental Kuznets curve hypothesis across development levels of Chinese provinces. Journal of Cleaner Production, 285, 124898. Crossref, Google Scholar
Al-Kaabi, A, H Al-Sulaiti, T Al-Ansari and HR Mackey [2021] Assessment of water quality variations on pretreatment and environmental impacts of SWRO desalination. Desalination, 500, 114831. Crossref, Google Scholar
Ardia, D, K Bluteau and A Kassem [2021] A century of economic policy uncertainty through the French–Canadian lens. Economics Letters, 205, 109938. Crossref, Google Scholar
Baker, SR, N Bloom and SJ Davis [2016] Measuring economic policy uncertainty. The Quarterly Journal of Economics, 131(4), 1593–1636. Crossref, Google Scholar
Battamo, AY, O Varis, JP Giesy, BT Oba, MA Aborisade, P Sun, Y Yang and L Zhao [2022] A framework for assessing freshwater vulnerability along China’s Belt and Road Initiative: An exposure, sensitivity and adaptive capacity approach. Environmental Science & Policy, 132, 247–261. Crossref, Google Scholar
Beal, G and J Bohlen [1957] The diffusion process (Special Report N^∘ 18, Agricultural Experiment Station). Iowa State College. Google Scholar
Białkowski, J, HD Dang and X Wei [2022] High policy uncertainty and low implied market volatility: An academic puzzle? Journal of Financial Economics, 143(3), 1185–1208. Crossref, Google Scholar
Bloom, N [2014] Fluctuations in uncertainty. Journal of Economic Perspectives, 28(2), 153–176. Crossref, Google Scholar
Branscomb, LM [1997] Confessions of a Technophile. New York: Springer Science & Business Media. Google Scholar
Caragliu, A and CF Del Bo [2019] Smart innovative cities: The impact of Smart City policies on urban innovation. Technological Forecasting and Social Change, 142, 373–383. Crossref, Google Scholar
Change, I [2001] Climate change 2001: The scientific basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change, 2001. Google Scholar
Chen, X and T Li [2022] Diffusion of agricultural technology innovation: Research progress of innovation diffusion in Chinese agricultural science and technology parks. Sustainability, 14(22), 15008. Crossref, Google Scholar
Cheng, M and C Duan [2021] The changing trends of internal migration and urbanization in China: New evidence from the seventh National Population Census. China Population and Development Studies, 5(3), 275–295. Crossref, Google Scholar
Cheng, Z, N Wang, R Ouyang, H Wang and Z Song [2022] A content analysis of the strictest water resources management policy in China. Water Economics and Policy, 8(02), 2250011. Link, Google Scholar
Deng, L, J Yin, J Tian, Q Li and S Guo [2021] Comprehensive evaluation of water resources carrying capacity in the Han River Basin. Water, 13(3), 249. Crossref, Google Scholar
Dodgson, M and J Bessant [1997] Effective innovation policy: A new approach. Long Range Planning, 1(30), 143. Crossref, Google Scholar
Doerfliger, N, P-Y Jeannin and F Zwahlen [1999] Water vulnerability assessment in karst environments: A new method of defining protection areas using a multi-attribute approach and GIS tools (EPIK method). Environmental Geology, 39(2), 165–176. Crossref, Google Scholar
Feldman, MP and R Florida [1994] The geographic sources of innovation: technological infrastructure and product innovation in the United States. Annals of the Association of American Geographers, 84(2), 210–229. Crossref, Google Scholar
Fornell, C and DF Larcker [1981] Evaluating structural equation models with unobservable variables and measurement error. International Journal of Research in Marketing, 18(1), 39–50. Crossref, Google Scholar
Gernego, YO, O Dyba and L Petrenko [2019] Determinants of innovative activities conserning socio-economic growth. Financial and Credit Activity Problems of Theory and Practice, 3(30), 445–453. Crossref, Google Scholar
Grimaldi, R and A Grandi [2005] Business incubators and new venture creation: An assessment of incubating models. Technovation, 25(2), 111–121. Crossref, Google Scholar
Gunzler, D, T Chen, P Wu and H Zhang [2013] Introduction to mediation analysis with structural equation modeling. Shanghai Archives of Psychiatry, 25(6), 390. Google Scholar
Haile, B, D Mekonnen, J Choufani, C Ringler and E Bryan [2022] Hierarchical modelling of small-scale irrigation: Constraints and opportunities for adoption in sub-Saharan Africa. Water Economics and Policy, 8(01), 2250005. Link, Google Scholar
Hamouda, MA, MM Nour El-Din and FI Moursy [2009] Vulnerability assessment of water resources systems in the Eastern Nile Basin. Water Resources Management, 23(13), 2697–2725. Crossref, Google Scholar
Hamouda, MAA (2006). Vulnerability assessment of water resources systems in the Eastern Nile Basin to environmental factors. Institute of African Research and Studies, Cairo University. Google Scholar
Huang, J, Z Wang, Z Jiang and Q Zhong (2022). Environmental policy uncertainty and corporate green innovation: Evidence from China. European Journal of Innovation Management (ahead-of-print). Google Scholar
J. Acs, Z and DB Audretsch [1989] Patents as a measure of innovative activity. Kyklos, 42(2), 171–180. Crossref, Google Scholar
Jayasiri, MMJGCN, S Yadav, NDK Dayawansa and K Ingold [2022] Managing agricultural water resources: Addressing the complexity of innovation, social perspectives, and water governance in Sri Lanka. Irrigation and Drainage, 71(S1), 71–85. Crossref, Google Scholar
Jiang Zhongyi, KW [2006] Basic Methods of Mathematical Economics. China: Peking University Press. Google Scholar
Kahrl, F, D Roland-Holst and D Zilberman [2005] New horizons for rural reform in China: resources, property rights and consumerism. Agricultural and Resources Economics Update, 9(1), 11–15. Google Scholar
Kanger, L, FW Geels, B Sovacool and J Schot [2019] Technological diffusion as a process of societal embedding: Lessons from historical automobile transitions for future electric mobility. Transportation Research Part D: Transport and Environment, 71, 47–66. Crossref, Google Scholar
Karakaya, E, A Hidalgo and C Nuur [2014] Diffusion of eco-innovations: A review. Renewable and Sustainable Energy Reviews, 33, 392–399. Crossref, Google Scholar
Khanam, M and T Daim [2021] A market diffusion potential (MDP) assessment model for residential energy efficient (EE) technologies in the US. Renewable and Sustainable Energy Reviews, 144, 110968. Crossref, Google Scholar
Kyaw, K [2022] Effect of policy uncertainty on environmental innovation. Journal of Cleaner Production, 363, 132645. Crossref, Google Scholar
Li, G, Y Lei, J Ge and S Wu [2017] The empirical relationship between mining industry development and environmental pollution in China. International Journal of Environmental Research and Public Health, 14(3), 254. Crossref, Google Scholar
Li, K and Q Li [2022] Social embeddedness and agricultural technology diffusion from the perspective of scale differentiation — A case study from China. International Food and Agribusiness Management Review, 26(1), 1–16. Google Scholar
Li, Q [2022] The problem of the impact of the concept of Chinese Ecological Civilisation upon environmental policy in the contemporary China. International Journal of Frontiers in Sociology, 4(8), 72–84. Google Scholar
Lu, Z-N, H Chen, Y Hao, J Wang, X Song and TM Mok [2017] The dynamic relationship between environmental pollution, economic development and public health: Evidence from China. Journal of Cleaner Production, 166, 134–147. Crossref, Google Scholar
Nesta, L, F Vona and F Nicolli [2014] Environmental policies, competition and innovation in renewable energy. Journal of Environmental Economics and Management, 67(3), 396–411. Crossref, Google Scholar
Noailly, J [2012] Improving the energy efficiency of buildings: The impact of environmental policy on technological innovation. Energy Economics, 34(3), 795–806. Crossref, Google Scholar
O’Callaghan, P [2020] Dynamics of water innovation: insights into the rate of adoption, diffusion and success of emerging water technologies globally. Wageningen University and Research. Google Scholar
Palm, A [2022] Innovation systems for technology diffusion: An analytical framework and two case studies. Technological Forecasting and Social Change, 182, 121821. Crossref, Google Scholar
Palmquist, R (1993). Ground water vulnerability: A DRASTIC approach. Air and Waste Management Association Annual Meeting, Vol. 15, pp. 91–150. Google Scholar
Qu, J, H Wang, K Wang, G Yu, B Ke, H-Q Yu, H Ren, X Zheng, J Li and W-W Li [2019] Municipal wastewater treatment in China: Development history and future perspectives. Frontiers of Environmental Science & Engineering, 13, 1–7. Crossref, Google Scholar
Rahman, MM and K Alam [2021] Clean energy, population density, urbanization and environmental pollution nexus: Evidence from Bangladesh. Renewable Energy, 172, 1063–1072. Crossref, Google Scholar
Robins, L, T Burt, L Bracken, J Boardman and D Thompson [2017] Making water policy work in the United Kingdom: A case study of practical approaches to strengthening complex, multi-tiered systems of water governance. Environmental Science & Policy, 71, 41–55. Crossref, Google Scholar
Rustico, E and S Dimitrov [2022] Environmental taxation: The impact of carbon tax policy commitment on technology choice and social welfare. International Journal of Production Economics, 243, 108328. Crossref, Google Scholar
Schumacker, RE and RG Lomax [2004] A Beginner’s Guide to Structural Equation Modeling. London: Psychology Press. Crossref, Google Scholar
Singh, S, T Jaakkola, ML Littman and C Szepesvári [2000] Convergence results for single-step on-policy reinforcement-learning algorithms. Machine Learning, 38, 287–308. Crossref, Google Scholar
Sun, Q, W Qi and X Yu [2021] Impacts of land use change on ecosystem services in the intensive agricultural area of North China based on Multi-scenario analysis. Alexandria Engineering Journal, 60(1), 1703–1716. Crossref, Google Scholar
Thompson, Jr. BH [1993] Institutional perspectives on water policy and markets. California Law Review, 81, 671. Crossref, Google Scholar
Tian, P, H Lu, J Liu, K Feng, R Heijungs, D Li and X Fan [2022] The pattern of virtual water transfer in China: From the perspective of the virtual water hypothesis. Journal of Cleaner Production, 346, 131232. Crossref, Google Scholar
Verschuur, J, EE Koks and JW Hall [2021] Global economic impacts of COVID-19 lockdown measures stand out in high-frequency shipping data. PLoS One, 16(4), e0248818. Crossref, Google Scholar
Wang, G, C Xiao, Z Qi, F Meng and X Liang [2021] Development tendency analysis for the water resource carrying capacity based on system dynamics model and the improved fuzzy comprehensive evaluation method in the Changchun city, China. Ecological Indicators, 122, 107232. Crossref, Google Scholar
Wang, X, L Liu, S Zhang and C Gao [2022] Dynamic simulation and comprehensive evaluation of the water resources carrying capacity in Guangzhou city, China. Ecological Indicators, 135, 108528. Crossref, Google Scholar
Wang, Y, Q Han, B De Vries and J Zuo [2016] How the public reacts to social impacts in construction projects? A structural equation modeling study. International Journal of Project Management, 34(8), 1433–1448. Crossref, Google Scholar
Wei, M, S Huang, L Li, T Zhang, W Akram, Z Khatoon and FG Renaud [2022] Evolution of water quality and biota in the Panjiakou Reservoir, China as a consequence of social and economic development: implications for synergies and trade-offs between sustainable development goals. Sustainability Science, 17(4), 1385–1404. Crossref, Google Scholar
Wei, S, K Lin, L Huang, Z Yao, X Bai and Z Chen [2022] Assessing the vulnerability of water resources system using VSD-SD coupling model: A case of Pearl River Delta. Water, 14(7), 1103. https://www.mdpi.com/2073-4441/14/7/1103 Crossref, Google Scholar
Wiser, RH and SJ Pickle [1998] Financing investments in renewable energy: The impacts of policy design. Renewable and Sustainable Energy Reviews, 2(4), 361–386. Crossref, Google Scholar
Xie, L and X Lu [2022] Circular economy and eco-industrial park. In Suzhou Industrial Park, pp. 33–59. New York: Springer. Crossref, Google Scholar
Yang, F, H Yuan and N Yi [2022] Natural resources, environment and the sustainable development. Urban Climate, 42, 101111. Crossref, Google Scholar
Yang, G, S Li, H Wang and L Wang [2022] Study on agricultural cultivation development layout based on the matching characteristic of water and land resources in North China Plain. Agricultural Water Management, 259, 107272. Crossref, Google Scholar
Ye, H [2022] The impact of knowledge spillovers on regional innovation performance: From the perspective of patent cooperation. World Scientific Research Journal, 8(3), 102–115. Google Scholar
Ye, J, Q Wan, R Li, Z Yao and D Huang [2022] How do R&D agglomeration and economic policy uncertainty affect the innovative performance of Chinese high-tech industry? Technology in Society, 69, 101957. Crossref, Google Scholar
Yuan, T, JG Wu, N Qin and J Xu [2022] Being nice to stakeholders: The effect of economic policy uncertainty on corporate social responsibility. Economic Modelling, 108, 105737. Crossref, Google Scholar
Yuan Yuan, ZY [2022] Progress and future prospects of water resources vulnerability at home and abroad. Journal of Arid Land resources and Environment, 36(7), 116–125. Google Scholar
Yunjiang, Y and G Xiangdong [2017] Market potential and wage difference of floating population: An analysis based on the perspective of heterogeneity. World Economy, 12, 98–118. Google Scholar
Zhai, Y, T Zhang, Y Bai, C Ji, X Ma, X Shen and J Hong [2021] Energy and water footprints of cereal production in China. Resources, Conservation and Recycling, 164, 105150. Crossref, Google Scholar
Zhang, C-Y and T Oki [2023] Water pricing reform for sustainable water resources management in China’s agricultural sector. Agricultural Water Management, 275, 108045. Crossref, Google Scholar
Zhang, C, V Lesser and P Shenoy [2009] A multi-agent learning approach to online distributed resource allocation. Twenty-first International Joint Conference on Artificial Intelligence. Google Scholar
Zhang, N [2020] Four Studies on Antitrust Enforcement and Foreign Investment Activities. University of Illinois at Urbana-Champaign. Google Scholar
Zhang, Y, J Deng, B Qin, G Zhu, Y Zhang, E Jeppesen and Y Tong [2022] Importance and vulnerability of lakes and reservoirs supporting drinking water in China. Fundamental Research, 3(2), 265–273. Google Scholar
Zhu, X., Zhao, K. and Zhu, D. [2015] The research on the problems relevant for water resources in coal mines. In 2015 3rd International Conference on Advances in Energy and Environmental Science. Atlantis Press, pp. 1433–1439. Crossref, Google Scholar
Zhu, X, K Zhao and D Zhu [2015] The research on the problems relevant for water resources in coal mines. 2015 3rd Int. Conf. Advances in Energy and Environmental Science, Atlantis Press, pp. 1433–1439. Crossref, Google Scholar