The Effect of Marine Pastures on Green Aquaculture in China

Abstract

Under the double pressures of economic growth and ecological environment protection, sea green transformation and the sustainable development of mariculture are critical. This paper constructs an evolutionary game model with the government as the main body and mariculture farmers (enterprises) as the main body and puts forward the research hypothesis. Based on 2006–2019 longitudinal data of nine provinces along China’s coast, using multi-period Difference-in-Difference (DID) and dual robust estimation, we empirically investigate the national oceanic ranch demonstration zones for the influence of the green sea aquaculture and their mechanism of action. The results showed that (1) the efficiency of green level of mariculture industry in China is not high, and the establishment of national marine pasture demonstration zone has not effectively promoted the improvement of green level of mariculture industry; (2) the institutional environment, unreasonable industrial structure, and lack of scientific and technological innovation have an effect on the national oceanic ranch demonstration area as the main causes of failure to effectively promote marine green farming; (3) the establishment of the national multi-period demonstration area in the north significantly hindered the growth of the green level of mariculture and fell into the “policy trap”, while the establishment of the national multi-period demonstration area in the south significantly promoted the growth of the green level of mariculture. The conclusions of this paper provide an empirical basis and reference for the improvement of the national marine pasture demonstration zone policy and the green transformation of mariculture to a certain extent.

1. Introduction

For many years, the mariculture model with mariculture as the main body has made great contributions to the growth of the marine economy and the development of related industries in China. Under the trend of increasingly scarce fishing resources, mariculture plays an indispensable role in ensuring national food security and improving the national dietary structure [1,2]. However, while China’s mariculture has made remarkable achievements, there are also a series of environmental and resource problems, such as the irrational utilization of mariculture resources [3,4,5], reduction in mariculture area [6,7,8], and increasing carbon emissions and nitrogen and phosphorus pollutant emissions from mariculture [9]. Vigorously promoting the green development of mariculture in China is not only of great significance to the high-quality development of China’s marine economy, but also has a profound impact on the realization of the dual carbon goal. In order to deal with a series of environmental and resource problems in China’s mariculture and accelerate the green development of mariculture, in 2017, the Ministry of Agriculture issued the “National Marine Pasture Demonstration Zone Construction Plan (2017–2025)”, which proposed that marine pasture aims to improve the marine ecological environment and promote marine fisheries towards green, coordinated, and sustainable development. In 2021, two sessions were authorized to issue the 14th Five-Year Plan for National Economic and Social Development of the People’s Republic of China and the Outline of 2035 Vision Goals, which proposed the need to accelerate the construction of marine pastures and promote the green development of mariculture. It is worth noting that whether it is the macro level of carbon sequestration and emission reduction, the growth of green farming economy, or the micro level of ecological quality improvement and the improvement of people’s happiness of life, the green transformation of mariculture plays a crucial role in promoting progress. Thus, we ask the following: Has the establishment of the national marine pasture demonstration zone promoted the green aquaculture of sea water? What is the mechanism of the establishment of the national marine pasture demonstration area affecting the green level of mariculture? Is there a difference between the establishment of national marine pasture demonstration zones in the south and the north on the impact of marine green aquaculture? The investigation and solution of the above series of problems are of great significance to the implementation of the national marine pasture demonstration zone policy and the acceleration of the green transformation of mariculture.

Multi-period analysis is an important part of the fishing economy. Modern multi-period analysis uses big data, artificial intelligence, digital platforms, and other technologies to conduct comprehensive monitoring, data collection, and data prediction of a multi-period [10]. In addition, the marine pasture is usually considered as a closed system, whose main feature is human intervention, and whose main purpose is the scientific and rational maintenance and utilization of human-oriented marine resources, so it can be regarded as an artificial closed system [11]. At present, the relevant research studies on marine pasture in China are gradually becoming mature, from the hard environment to the soft environment, from the site selection of marine pasture [12,13], properties of algal reef materials [14,15], and spatial layout [16,17] to modern marine pasture construction [15], ecological security assessment [18,19,20], and research on the key technologies of marine pasture [21,22]; from the emphasis on economic benefits to the consideration of both economic and ecological benefits, the analysis of potential economic benefits [23,24] to ecological benefits [25,26], and the measurement of comprehensive benefits [27,28]. In recent years, with the “two mountains theory” and the “two-carbon target” successively proposed, the academic community has focused on topics such as carbon sinks in marine pastures and fisheries [29,30], most scholars have measured the blue carbon sink with marine shellfish and algae as objects [31,32], and some scholars have also started from carbon sink pricing and incentive subsidies [33,34].

Marine green aquaculture, also known as “marine carbon sink fisheries”, refers to the reduction in the amount of dissolved carbon dioxide in seawater through the carbon sequestration capacity of marine organisms themselves, thus affecting the marine carbon cycle activities, processes, and mechanisms of fishery production activities [35]. In addition, marine green aquaculture can also be understood as promoting green technology innovation and mechanism and system innovation through the formulation of green development plans and the establishment of standards for marine aquaculture fisheries, so as to achieve a harmonious coexistence of people, aquaculture production activities, and the marine ecological environment [36]. The measurement of the green level of mariculture and its effect are research hotspots of mariculture in China. As for the measurement and evaluation of the green level of mariculture, two main methods are used at present. The first is to measure and evaluate the green level of mariculture from a macro perspective by using various weighting methods [37] or an ecological dynamic system model [38]. The second is data enveloping analysis from a microscopic perspective [39,40,41]. At present, the effect of the green level of mariculture comes mainly from the perspective of environmental regulation. By building an evolutionary game model, some scholars found that government environmental regulations greatly affected the production behavior of marine green aquaculture by fishermen (households), and appropriate regulation intensity improved the level of marine green aquaculture to a certain extent [42]. By building a regression model, some other scholars found that the influence of different types of environmental regulations on the green level of mariculture is non-linear [43].

However, most of the literature on the relationship between marine ranches and marine green aquaculture consists of qualitative studies, which mainly provide a theoretical basis for the green transformation of marine aquaculture from the aspects of breeding variety selection, structural improvement, and the improvement of aquaculture technology. Papageorgiou et al. [44] proposed that mariculture varieties with a strong carbon sequestration capacity should be selected and scientifically combined to ensure the sustainable development of mariculture. Long et al. [45] proposed that advanced technologies should be introduced to effectively maintain the marine ecological environment in the process of mariculture production. In summary, it can be found that there are few literature articles which evaluate the policy effect of marine pasture, and there is a lack of studies on the impact of marine pasture on marine green aquaculture.

Compared with the existing literature, the possible marginal contributions of this paper are as follows: First, it tested the influence of the establishment of national marine pasture demonstration zones on the green level of mariculture and enriched the relevant research in the field of the policy evaluation of national marine pasture demonstration zones. Second, the evolutionary game model was innovatively used to analyze the internal mechanism of the influence of the establishment of national marine pasture demonstration zones on the green level of mariculture, and the mechanism was clarified from three aspects: institutional environment, industrial structure, and scientific and technological innovation. Third, against the background of the establishment of national marine pasture demonstration zones in batches, the standard DID is no longer applicable, so multi-period DID is used for a preliminary test, and then the more advanced dual-robust estimation of multi-period DID is used for a robustness test, so as to obtain a more accurate net effect of the policy. Fourth, the triple difference method was used to test the effect mechanism of the establishment of national marine pasture demonstration zones on the green level of mariculture, and the heterogeneity of the effect of the establishment of national marine pasture demonstration zones on mariculture was tested with the help of multi-period DID and its double robust estimators, in order to provide a reference for the green transformation of marine aquaculture and the construction of marine pastures in the north and south according to local conditions.

2. Theoretical Analysis and Research Hypothesis

2.1. Mechanism Analysis of the Influence of the Establishment of National Marine Pasture Demonstration Zones on the Green Level of Mariculture

This paper uses an evolutionary game model to deduce the influence of national marine pasture demonstration zones on the green level of mariculture, based on two aspects: First, from the perspective of combining theory with practice, the behavior choice between the government and mariculture farmers (enterprises) basically conforms to the idea of an evolutionary game. There is inevitably a contradiction between the extensive growth of economic output value in exchange for resources and the green transformation of mariculture. By establishing national marine pasture demonstration zones to regulate the production of mariculture farmers (enterprises), the production capacity of mariculture farmers (enterprises) will be limited to a certain extent, affecting the growth of single economic output value. Therefore, by constructing an evolutionary game model between the government and mariculture farmers (enterprises) and studying the equilibrium state of the strategies used by these two interest groups between the policies of the national marine pasture demonstration area and the green transformation of mariculture, the game evolution law between the government’s national marine pasture demonstration area policies and the mariculture green transformation of mariculture by mariculture farmers (enterprises) can be revealed. Second, from the perspective of economic tools, an evolutionary game is an analytical model that combines game theory and dynamic evolution process, which can analyze the evolutionary dynamic selection process of the competition and balance relationship between limited rational agents in a system and solve the behavioral choice problem of the government and mariculture farmers (enterprises).

From the perspective of the government, it is assumed that the probability of strong policy implementation in the national marine pasture demonstration zone is x, and the probability of weak policy implementation in the national marine pasture demonstration zone is 1 − x. The probability of mariculture farmers (enterprises) making the green transformation of mariculture is y, and the probability of maintaining the status quo is 1 − y; then, there is $x,y\in \left[0,1\right]$. In the process of implementing the national marine pasture demonstration area policy, the main benefits and costs considered by the government include the baseline income E₁; when the national marine ranch demonstration zone policy implementation is strong, its environmental benefit is E₂; the economic benefit loss caused by capacity reduction is L₁; the additional environmental loss caused by mariculture farmers (enterprises) to maintain the status quo is L₂; and the implementation cost is C₁. When the implementation of the national marine pasture demonstration zone policy is weak, the environmental loss caused by the maintenance of the status quo by mariculturists (enterprises) is L₃ and the implementation cost is C₂. In the process of mariculture green transformation, the main benefits and costs considered by mariculture farmers (enterprises) include the baseline income E₃; in the case of mariculture green transformation, the transformation cost is C₃; in the case of a national marine pasture demonstration zone, the additional reputation and incentive benefits brought by the active transformation are E₄ when the policy implementation is weak. In the case of maintaining the status quo, the economic loss of the national marine pasture demonstration zone is L₄ when the policy implementation is strong, and the additional economic loss of the national marine pasture demonstration zone is L₅ when the policy implementation is weak. According to the above conditions, the game matrix between the government and mariculture farmers (enterprises) can be obtained, as shown in

Table 1. Game matrix between the government and mariculture farmers (enterprises).

		Mariculture Farmer (Enterprise)
		Green Transformation of Mariculture (y)	Maintain the Status Quo (1 − y)
Government	The implementation of the national marine pasture demonstration zone policy is strong (x)	(E1 + E2 − C1 − L1, E3 − C3)	(E1 − C1 − L2, E3 − L4)
	The implementation of the national marine pasture demonstration zone policy is weak (1 − x)	(E1 − C2, E3 − C3 + E4)	(E1 − C2 − L3, E3 − L5)

Note: C₁ > C₂ > 0, L₂ > L₃ > 0, L₄ > L₅ > 0, E₄ ≥ 0, E₁ > 0, E₂ > 0, E₃ > 0, L₁ > 0, C₃ > 0.

.

According to Table 1, the expected return (Er_y) of mariculture green transformation and the expected return (Er_1−y) of maintaining the status quo can be obtained, as shown in Equation (1).

$$\{\begin{array}{l}E{r}_y=x(E_3-C_3)+(1-x)(E_3-C_3+E_4)\\ E{r}_{1-y}=x(E_3-L_4)+(1-x)(E_3-L_5)\end{array}$$

(1)

The average expected benefits are

$$\overline{Er}=yE{r}_y+(1-y)E{r}_{1-y}$$

(2)

Then, the dynamic equation of replication is

$$\begin{array}{l}F(y)=\frac{dy}{dt}=y(E{r}_y-\overline{Er})=y(1-y)(E{r}_y-E{r}_{1-y})\\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}=y(1-y)(x{L}_4-x{L}_5-x{E}_4+L_5+E_4-C_3)\end{array}$$

(3)

The first derivative of F(y) according to Equation (3) is

$$\hspace{1em}{F}^{\prime }(y)=(1-2y)(x{L}_4-x{L}_5-x{E}_4+L_5+E_4-C_3)$$

(4)

When $F^{\prime }(y)=0$ reaches the maximum expected value, we obtain

$$x=\frac{E_4+L_5-C_3}{E_4+L_5-L_4}$$

(5)

According to the stability theorem of the differential equation, $F(y)=0$ and $F^{\prime }(y)<0$ must be satisfied for the unilateral strategy of the mariculture farmer (enterprise) to be in a stable state. When x = (E₄ + L₅ − C₃)/(E₄ + L₅ − L₄), $x=x^{\ast }$, $F(y)=0$, and $F^{\prime }(y)=0$, this indicates that when the probability of strong implementation of the national marine pasture demonstration zone policy is $x^{\ast }$, mariculturists (enterprises) can achieve the maximum expectation regardless of whether they carry out the green transformation of mariculture. At this time, $y\in \left[0,1\right]$ is stable, and it is impossible to judge the stability strategy of mariculture farmers (enterprises). When x ≠ (E₄ + L₅ − C₃)/(E₄ + L₅ − L₄), it is necessary to discuss the case of E₄ + L₅ − C₃ and E₄ + L₅ − L₄.

Case 1: If 0 < E₄ + L₅ − C₃ < E₄ + L₅ − L₄, the following two scenarios should be discussed: when $x>x^{\ast }$, $F^{\prime }(y=0)<0$ and $F^{\prime }(y=1)>0$ can be obtained, then $y=0$ is stable. At this time, the stability strategy of mariculturists (enterprises) is to maintain the status quo, which is the result of the increase in production costs of mariculturists (enterprises) and the gradual disappearance of extra reputation and incentive benefits during the transformation. When $x<x^{\ast }$, $F^{\prime }(y=0)>0$, and $F^{\prime }(y=1)<0$ can be obtained, then $y=1$ is stable. At this time, the stability strategy of mariculture farmers (enterprises) is the green transformation of mariculture, which means that for mariculture farmers (enterprises) to obtain a good social image, the extra reputation and incentive benefits of transformation are higher than the transformation cost.

Case 2: If 0 > E₄ + L₅ − C₃ > E₄ + L₅ − L₄, the following two scenarios need to be discussed: when $x>x^{\ast }$, $F^{\prime }(y=0)>0$, and $F^{\prime }(y=1)<0$ can be obtained, then $y=1$ is stable. At this time, the stable strategy of mariculture farmers (enterprises) is the green transformation of mariculture. When $x<x^{\ast }$, $F^{\prime }(y=0)<0$, and $F^{\prime }(y=1)>0$ can be obtained, then $y=0$ is stable. At this time, the stabilization strategy of mariculturists (enterprises) is to maintain the status quo. In this case, the government and mariculture farmers (enterprises) will constantly adjust the game strategy, and there is no stable equilibrium point.

Case 3: If E₄ + L₅ − C₃ < E₄ + L₅ − L_4, E₄ + L₅ − C₃ < 0, and E₄ + L₅ − C₃ < E₄ + L₅ − L₄ < 0, it always exists that $x<x^{\ast }$, we can obtain $F^{\prime }(y=0)<0$ and $F^{\prime }(y=1)>0$, and then $y=0$ is stable. At this time, the stabilization strategy of mariculture farmers (enterprises) is to maintain the status quo. If E₄ + L₅ − C₃ < 0 < E₄ + L₅ − L₄, it always has $x>x^{\ast }$, we can obtain $F^{\prime }(y=0)<0$ and $F^{\prime }(y=1)>0$, and then $y=0$ is stable. At this time, the stabilization strategy of mariculturists (enterprises) is to maintain the status quo. In this case, no matter whether the policy implementation of the national marine pasture demonstration zone is strong or weak, the benefit of mariculturists (enterprises) choosing transformation is lower than that of maintaining the status quo, and the best strategy for mariculturists (enterprises) is to maintain the status quo.

Case 4: If E₄ + L₅ − C₃ > E₄ + L₅ − L_4, E₄ + L₅ − C₃ > 0, and E₄ + L₅ − C₃ > 0 > E₄ + L₅ − L₄, it always exists that $x>x^{\ast }$, we can obtain $F^{\prime }(y=0)>0$ and $F^{\prime }(y=1)<0$, and then $y=1$ is stable. At this time, the stable strategy of mariculture farmers (enterprises) is the green transformation of mariculture. If E₄ + L₅ − C₃ > E₄ + L₅ − L₄ > 0, it always has $x<x^{\ast }$, we obtain $F^{\prime }(y=0)>0$ and $F^{\prime }(y=1)<0$, and then $y=1$ is stable. At this time, the stable strategy of mariculture farmers (enterprises) is the green transformation of mariculture. In this case, no matter whether the policy implementation of the national marine pasture demonstration area is strong or weak, the benefits of mariculture farmers (enterprises) choosing to transform are higher than the benefits of maintaining the status quo, and the best strategy is for mariculture farmers (enterprises) to carry out the green transformation of mariculture.

Based on the game results of the above four situations, it can be found that if there are immature market conditions such as the reputation mechanism, incentive mechanism, and supervision mechanism for the green transformation of mariculture, mariculture farmers (enterprises) will choose to maintain the status quo even if the policy implementation of the national marine pasture demonstration zone is strong. Therefore, in the process of implementing the national marine pasture demonstration zone policy, it is necessary to gradually improve the reputation, incentive, and supervision mechanism, otherwise it will increase the difficulty for mariculture farmers (enterprises) to carry out the green transformation of mariculture. Accordingly, the following hypothesis is proposed in this paper:

H1. 

Under the background of immature market conditions such as reputation mechanisms, incentive mechanisms, and supervision mechanisms, the national marine pasture demonstration zone policy has not effectively promoted marine green aquaculture.

2.2. Mechanism Analysis of the Influence of the Establishment of National Marine Pasture Demonstration Zones on the Green Level of Mariculture

Above, an evolutionary game model between the government and mariculture farmers (enterprises) is constructed under the background of introducing reputation, incentive, and supervision mechanisms. The study found that the national marine pasture demonstration zone policy may not effectively promote the growth of the green level of mariculture. The causes of this phenomenon will be discussed in the following three aspects: institutional environment, industrial structure, and scientific and technological innovation. From the perspective of institutional environment, the “National Marine Pasture Demonstration Zone Construction Plan (2017–2025)” clearly points out that there is a lack of institutional mechanism construction in China’s marine pasture. China’s marine pasture started late, and the relevant supervision, laws, and systems are not perfect [19,46]; they are prone to the problem of unclear rights and responsibilities among the subjects of marine pasture construction, operation, and supervision. There is a phenomenon of “heavy construction, light management” in marine pastures in some areas, which is prone to the problem of inadequate supervision, which provides a breeding ground for the one-sided pursuit of economic benefits and short-term interests. From the perspective of industrial structure, on the one hand, the industrialization degree of China’s national marine pasture demonstration area is low, the construction, development, and utilization of the national marine pasture demonstration area focus on the primary industry, and the development of the second and third industries is insufficient, resulting in a weak industrial cluster degree and a low integration degree of the three industries [19]. On the other hand, the foundation of intelligent equipment for marine pasture is weak [47]. If the ocean ranch is blindly invested in the early stage of construction, there will be problems of intellectualization and low-end locking of equipment and facilities, which will lead to the optimization of the industrial structure. From the perspective of scientific and technological innovation, on the one hand, the current national marine pasture demonstration zone does not pay enough attention to scientific and technological innovation [48], and the national marine pasture demonstration zone has a low degree of science and technology, key technologies, and systemic research, and development challenges remain to be overcome [49]. On the other hand, with the construction and development of national marine pasture demonstration zones, the research and development and application of core technologies have become more and more important. However, compared with the rapidly developing national marine pasture demonstration zones, the progress of technology research and development and achievement transformation related to marine pasture is relatively slow; that is, scientific and technological innovation is relatively lagging behind [50]. Accordingly, the following hypothesis is proposed in this paper:

H2. 

A poor institutional environment, unreasonable industrial structure, and insufficient scientific and technologicalinnovation are the main reasons for the failure of national marine pasture demonstration zone policies to effectively promote the growth of the green level of mariculture.

2.3. Heterogeneity Analysis of the Impact of the Establishment of National Marine Pasture Demonstration Zones on the Green Level of Mariculture

Clarifying the difference between the national marine pasture demonstration zones in the south and the north in their impact on the green level of mariculture will have a profound impact on building a “modern upgraded version” of marine pasture in the north and expanding the “strategic new space” of marine pasture in the south [19]. From the perspective of marine pasture types, it is clearly pointed out in the Construction Plan of the National Marine Pasture Demonstration Zone (2017–2025) that the Yellow and Bohai Sea areas in China are mainly dominated by breeding and leisure marine pastures, the East Sea area is mainly dominated by leisure and conservation marine pastures, and the South Sea area is mainly dominated by conservation marine pastures. It can be seen that the northern marine pastures pay more attention to breeding and leisure functions, while the southern marine pastures pay more attention to conservation functions. Therefore, compared with the northern marine pastures, the southern marine pastures have a better foundation and greater advantages in terms of ecological conservation. From the perspective of the scale of conservation marine pastures, the northern coastline of China is relatively flat, and there are good conditions for building large-scale offshore marine pastures, so the ecological restoration marine pastures are mostly distributed offshore. However, in terms of target positioning, marine pastures in the north pay more attention to breeding and leisure and invest less in conservation marine pastures. Therefore, marine pastures in the north are mainly small and medium-sized coastal ecological restoration marine pastures. The coastline of the southern sea of China is tortuous, the sea and land are not coherent, and the main positioning of the southern sea pasture is of the conservation type, so the southern region is mainly dominated by large and medium-sized offshore ecological restoration marine pastures. Therefore, it is easier to realize the scale effect of ecological conservation in the southern marine pasture. From the perspective of the specific operation of marine ranches, the initial cost input and artificial reef input of marine ranches in northern China are unreasonable, and the operation and management bodies are mostly enterprises, which are more likely to produce economic benefits rather than ecological benefits [51]. On the other hand, the model of the proliferation, release, and environmental restoration of marine pastures in the south is more likely to bring ecological benefits and realize intensive management [19]. Accordingly, the following hypothesis is proposed in this paper:

H3. 

The establishment of the national marine pasture demonstration zone in northern China failed to promote the growth of the green level of mariculture and fell into the “policy trap”. The national marine pasture demonstration zone in southern China has promoted green aquaculture in seawater.

3. Research Design

3.1. Research Methods and Model Construction

3.1.1. Super-SBM Model

In this paper, the Super-Slacks Based Measure (Super-SBM) is used to measure the green level of mariculture. The Super-SBM model is a modified model for relaxation variables of the SBM model proposed by Tone [52], which solves the problem that the Slacks Based Measure (SBM) cannot evaluate and rank effective decision-making units (DMUs). The model is described as follows:

The data of 9 coastal provinces in China, namely 9 DMU samples, were selected. Each DMU contains three parts: input, expected output, and unexpected output, which are represented by x, y, and z, respectively. Each province has a kind of input in the process of mariculture, which is denoted as $x_a^i$ and represents the input value of production unit i in year j; b is the expected output, let us call it $y_b^i$, which represents the expected output value of production unit i in year j; type c is an undesirable output, denoted as $z_c^i$, which represents the undesired output value of production unit i in year j. The green level of mariculture of DMU in each province is measured by solving the following model:

$$\begin{array}{l}{\phi }^{\ast }=\min \frac{1-\left(\frac{1}{A}{\displaystyle \sum _{a=1}^A\frac{s_a^x}{{x_a^i}^{\prime }}}\right)}{1+\left[\frac{1}{B+1}\left({\displaystyle \sum _{b=1}^B\frac{s_b^y}{{y_b^i}^{\prime }}}\right)+{\displaystyle \sum _{z=1}^Z\frac{s_c^z}{{z_c^i}^{\prime }}}\right]}\\ s.t.\{\begin{array}{l}{\displaystyle \sum _{i=1}^I{\lambda }_i^x{x}_a^i-s_a^x={x_a^i}^{\prime },a=1,\dots ,A\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}}\\ {\displaystyle \sum _{i=1}^I{\lambda }_i^y{y}_b^i-s_b^y={y_b^i}^{\prime },b=1,\dots ,B}\\ {\displaystyle \sum _{i=1}^I{\lambda }_i^y{z}_c^i-s_c^z={z_c^i}^{\prime },c=1,\dots ,C}\\ {\lambda }_i^x\ge 0,s_a^x\ge 0,s_b^y\ge 0,s_c^z\ge 0,i=1,2,\dots ,I\end{array}\end{array}$$

(6)

In Formula (6), ${\phi }^{\ast }$ represents the efficiency value. If the value is greater than 1, the effective DMU of SBM can be evaluated and sorted. ${\lambda }_i^x$, ${\lambda }_i^y$ respectively represent the weight of each input value and output value of the production unit; $s_a^x$, $s_b^y$, and $s_c^z$ represent the relaxation variable.

3.1.2. Multi-Period DID and Its Double Robust Estimators

Since the establishment of the first batch of national marine ranch demonstration zones at the end of 2015, China has paid more attention to the construction of modern marine pastures and has issued a series of policies related to marine pastures, and by the beginning of 2022, China had set up a total of seven batches of national marine pasture demonstration zones. In this paper, the establishment of the national marine pasture demonstration zone is regarded as a “quasi-natural experiment”. In addition to the policy of the establishment of the national marine pasture demonstration zone, there are still many factors affecting marine green aquaculture. In addition, the establishment of the national marine pasture demonstration zone is carried out in batches. Therefore, multi-period DID was used to test the influence of the establishment of a national marine pasture demonstration area on marine green aquaculture.

In this paper, the provinces that have been approved to set up national marine pasture demonstration zones are taken as the treatment group, and the provinces that have not set up national marine pasture demonstration zones are taken as the control group. According to the time when each province was approved to set up a national marine pasture demonstration zone, the dummy variable Mp_it was set up. If the province was approved to set up a national marine pasture demonstration zone in the same year, then Mp_it = 1, otherwise Mp_it = 0. In addition, in the process of assigning values, this paper treats the establishment time of the national marine pasture demonstration zone as follows: If the province is approved to establish the national marine pasture demonstration zone in the first half of the year, the initial year of establishment is that year. If the national marine pasture demonstration zone is approved in the second half of the year, the initial year of its establishment will be the next year. In summary, the model is constructed as follows:

$$MG{L}_{it}={\beta }_0+{\beta }_{1}M{p}_{it}+\theta Control{s}_{it}+{\omega }_i+{\upsilon }_t+{\epsilon }_{it}$$

(7)

In Formula (7), i represents the province and t represents the year; MGL_it is the explained variable, which represents the green level of mariculture in year t of province i. β₁ is the core estimation parameter to measure the net effect of the establishment of national marine pasture demonstration zones on the green level of mariculture. If β₁ is significantly positive, it indicates that the establishment of national marine pasture demonstration zones promotes the growth of the green level of mariculture; otherwise, there is a hindering effect. Controls_it are control variables; ${\omega }_i$ stands for an individual fixed effect; ${\upsilon }_t$ represents a time fixed effect; ${\epsilon }_{it}$ is a random disturbance term.

Currently, most of the existing literature [53,54] adopts a two-way fixed effect model (TWFE) for multi-period DID analysis. However, in practical applications, TWFE’s two assumptions of the intertemporal homogeneity of processing effects and the independence of control variables are difficult to realize, resulting in bias problems. Therefore, in order to alleviate the possible bias in TWFE estimation of the effect of multi-period DID processing, this paper uses the double-robust estimator of multi-period DID to test the robustness of the multi-period DID regression results.

3.2. Variable Selection

3.2.1. Explained Variables

In this paper, the mariculture green level MGL_it was selected as the explained variable. In order to study the coordinated development of the green economy, environmental emission reduction, resource conservation, and ecological environment optimization of mariculture in China, the Super-SBM model was used to measure the green level of mariculture. The index system is shown in

Table 2. Index system of mariculture green level measurement.

Primary Index	Secondary Index	Variable	Instructions
Input	Resources	Fixed assets per unit of breeding area	Marine motor fishing vessels (production fishing vessels) at the end of the year/mariculture area
		Farming area per unit of labor force	Mariculture area/number of mariculture professionals
		Fishery seedlings per unit of cultivation area	Number of mariculture seedlings/mariculture area
	Labor	Labor force per unit of breeding area	Number of professionals in mariculture/number of professionals in fisheries
	Technology	Technical training intensity	number of fishermen technical training × number of mariculture professional practitioners/number of fishery aquaculture professional practitioners.
Expected output	Economy	Economic output per unit of labor	Mariculture output value/number of mariculture professionals
	Ecology	Carbon sequestration per unit of farming area	Carbon sequestration in mariculture/mariculture area
Undesirable output	Ecological pollution	Nitrogen and phosphorus pollution per unit of farming area	Nitrogen and phosphorus pollution output/mariculture area
		Carbon emissions per unit of farming area	Mariculture carbon emissions/mariculture area

. The selection process of the indicators is as follows:

The input index reflects the input of resources, labor, and technical elements in mariculture practice activities. Based on the practice of Zhang et al. [55], the mariculture fixed assets per unit aquaculture area, mariculture fishery seedlings, mariculture labor force, aquaculture area per unit labor force, and technical training intensity are selected as input indicators for the green level of mariculture. The calculation formula of the technical training intensity is as follows: technical training intensity = number of fishermen technical training × number of mariculture professional practitioners/number of fishery aquaculture professional practitioners.

For the expected output index, this paper reflects the beneficial output of mariculture activities from both economic and ecological output. Based on the practice of Ren and Zeng [39], the economic output per unit of labor force and carbon sequestration per unit of aquaculture area were selected as the expected output indicators of the green level of mariculture. Based on the practice of Wu and Li [56], this paper calculates the carbon sequestration per unit of aquaculture area. The carbon sequestration amount generated in the process of shellfish and algae cultivation was calculated to obtain the carbon sequestration amount of mariculture, and then the ratio of the carbon sequestration amount of mariculture to mariculture area was used to characterize the carbon sequestration amount per unit aquaculture area.

In this paper, ecological pollution is used to reflect the negative output of mariculture activities. Drawing on the practice of Guo et al. [57], the nitrogen and phosphorus pollution per unit aquaculture area and carbon emissions per unit aquaculture area were selected as the non-expected output indicators of the green level of mariculture. Meanwhile, Sun et al. [43]’s method was used for reference to measure the amount of nitrogen and phosphorus pollution per unit of aquaculture area. From the two aspects of feeding culture and non-feeding culture, the nitrogen and phosphorus pollution yields were calculated and added together. Meanwhile, the ratio of the nitrogen and phosphorus pollution yield to the mariculture area was used to characterize the nitrogen and phosphorus pollution amount per unit aquaculture area. In addition, this paper draws on the practice of Guan et al. [58] to measure the carbon emissions per unit of farming area. Based on energy combustion and power consumption, the carbon emissions of mariculture were calculated and summed up. Finally, the ratio of the carbon emissions of mariculture to the mariculture area was used to characterize the carbon emissions per unit area of mariculture.

3.2.2. Core Explanatory Variables

According to the list of each batch of national marine pasture demonstration zones published by the Ministry of Agriculture and Rural Affairs, and the provinces that have been approved to establish national marine pasture demonstration zones at various periods of time, the core variable Mp_it of national marine pasture demonstration zones is constructed as the core explanatory variable of this paper.

3.2.3. Control Variables

In this paper, according to the index proposed by Sun et al. [43], the intermediate consumption (AIC), income level (FPI), and fishery drug use (AFM) of mariculture were selected as control variables. The green level of mariculture is not only related to the aquaculture activities themselves, but also closely related to resources, the economy, and the environment [59]. Firstly, resource input provides a material guarantee for the improvement of aquaculture efficiency and quality. In this paper, the intermediate consumption of mariculture is selected as the resource control variable, and the calculation formula is as follows: intermediate consumption of mariculture = intermediate consumption of fishery × total output value of mariculture/total output value of fishery. Secondly, economic development determines residents’ living standards, and changes in residents’ living standards will affect consumers’ demand for high-quality food, which will lead to changes in the yield structure of mariculture. Therefore, the income level of mariculture is selected as the economic control variable in this paper. Finally, environmental factors will directly affect the green level of marine aquaculture, and the rational use of fishing drugs has a positive impact on marine green aquaculture. Therefore, the use of fish drugs in mariculture was selected as the environmental control variable in this paper.

3.3. Data Sources and Descriptive Statistics

Based on the availability of data, this paper selected the longitudinal data of 9 provinces (due to the serious lack of data, the data of Shanghai and Tianjin are excluded) in China from 2006 to 2019 (Due to the serious lack of city and county data, this paper uses provincial longitudinal data for empirical analysis. In 2020, all coastal provinces set up national-level marine pasture demonstration zones, so the data were updated to 2019. If the data are updated after 2019, there will be no individuals in the control group who have not been affected by the policy, resulting in the failure to carry out the policy evaluation of the national marine pasture demonstration area). The data were selected to empirically test the influence and mechanism of the establishment of national marine pasture demonstration zones on the green level of mariculture. The original data of aquaculture came from the “China Fishery Statistical Yearbook”, the original data of the pollution production coefficient were selected from “the Manual of the First National Pollution Source Census of Aquaculture Pollution Source”, the original data of the carbon emission coefficient were drawn from “the IPCC National Greenhouse Gas Inventory Guide”, the original data of the fuel consumption conversion coefficient of marine fishing vessels were chosen from “the Domestic Motor Fishing Vessels”, and the original data of the conversion coefficient of “Energy (diesel and electricity)—Standard coal” are from the “China Energy Statistical Yearbook”. The definitions of the variables and descriptive statistics are shown in

Table 3. Variable description and descriptive statistics.

Variable	Calculation Method	Sample Size	Mean Value	Standard Deviation
Mariculture Green Level (MGLit)	The green level of mariculture as measured by the Super-SBM model	126	0.222	0.416
National Marine Pasture Demonstration Zone (Mpit)	Dummy variable (0, 1)	126	1.245	0.267
Intermediate Consumption (AIC)	Intermediate fishery consumption × total output value of mariculture/total output value of fishery	126	5.320	0.857
Income Level (FPI)	Per capita disposable income of fishermen after logarithmic treatment	126	9.485	0.389
Fisheries Use (AFM)	Output value of fishery medicine/total output value of mariculture	126	0.005	0.008
Fishing Technology Extension (FTP)	ln(1 + aquaculture technology extension funds/total aquaculture area)	126	0.001	0.001
Industrial Structure (TIS)	ln(1 + mariculture value/gross regional product)	126	0.011	0.009
Science and Technology Output (STO)	ln(1 + number of patents granted by marine research institutions)	126	3.502	1.993

.

4. Empirical Results and Analysis

4.1. Efficiency Measurement Results and Analysis

In this paper, MaxDEA Ultra 8 software was used to calculate the green level of mariculture in nine coastal provinces of China from 2006 to 2019 based on Formula (6) and the Super-SBM model. The results are shown in Figure 1 and Figure 2.

From the overall average point of view, the green level of mariculture in China is between 0.7 and 1.0, and the efficiency is not good, which indicates that the green level of mariculture in China is not high, and there is still a large space for improvement. From the perspective of time trends, the green level of mariculture in China showed a relatively stable trend; its value reached a peak in 2008 and then gradually fell, reached a low stage in 2011, and then fluctuated back up. This may be due to the fifth wave of mariculture in China, represented by abalone and sea cucumber seafood breeding, which was pushed to the peak in 2008. Abalone and sea cucumber need a clean and green breeding environment. In the process of large-scale abalone and sea cucumber farming, ecological conservation will be indirectly promoted and the development of green aquaculture will be promoted, thus boosting the green level of mariculture to reach a peak in 2008. Since then, with the expansion of mariculture scale, coupled with the relatively backward development of mariculture technology, supervision, and systems, problems such as the increased pollution of mariculture and unreasonable development and utilization of resources have become more prominent, resulting in the green level of mariculture falling to a low point in 2011. In 2015, a new development concept (Proposed by the Chinese leadership, it refers to promoting social and economic development, emphasizing innovative development, coordinated development, green development, open development and shared development) was put forward, and in the same year, China set up the first batch of national marine pasture demonstration areas, gradually opening the curtain of the sixth wave of mariculture. Under the guidance of the concept of green development, the green level of mariculture has gradually recovered, but it has not returned to the high point in 2008, which may be related to the failure of marine pastures to fully drive the green transformation of mariculture.

From 2006 to 2019, the mean value of the mariculture green level in China showed an unbalanced regional distribution, among which, the mean value of the mariculture green level in provinces in the Yellow Sea and Bohai Sea region had a small difference, while the mean value of the mariculture green level in provinces in the East China Sea and the South China Sea had a large difference. From the provincial level, Fujian in the East China Sea and Hainan in the South China Sea had a high level of green mariculture, and the two provinces were ahead of other coastal provinces in green mariculture. From the perspective of culture structure, Fujian is dominated by shellfish culture, with a large production scale and strong carbon sequestration ability. Hainan actively promotes the transformation of mariculture to shore and deep sea, and it has made strong efforts to clear and rectify offshore aquaculture, mature ecological dynamic monitoring and environmental observation and forecasting technologies, and strong ecological conservation and restoration capabilities. The green levels of mariculture in Jiangsu and Guangxi in the East China Sea region and the South China Sea region were lower, which lowered the green level of mariculture in the East China Sea region and the South China Sea region, respectively. This may be due to the large proportion of fish farming in the Jiangsu economy, the unreasonable structure of the marine fishery industry, and the dominant growth mode of extensive fishery; Guangxi marine green aquaculture started late, there are few technical achievements related to marine green aquaculture, and the corresponding technical achievements’ transformation and application efficiency are low, which leads to the slow progress of marine green aquaculture in Guangxi.

4.2. Benchmark Regression Results

In this paper, the national marine pasture demonstration zone policy is regarded as a quasi-natural experiment, and multi-period DID is used to test the policy effect of the establishment of national marine pasture demonstration zone on the green level of mariculture. The benchmark regression results are shown in

Table 4. Benchmark regression results.

	(1)	(2)	(3)	(4)
Variable	MGL	MGL	MGL	MGL
Mpit	−0.028 (−0.280)	−0.022 (−0.215)	−0.037 (−0.344)	−0.042 (−0.384)
AFM		−6.878 (−0.843)	−9.672 (−1.682)	−11.760 * (−2.076)
FPI			0.429 (1.169)	0.339 (0.864)
AIC				0.000 (0.598)
Constants	1.326 *** (13.024)	1.359 *** (11.805)	−2.901 (−0.793)	−2.141 (−0.553)
Province FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
N	126	126	126	126
R2	0.024	0.023	0.062	0.064

Notes: The t value is in parentheses. *** represents that the level of 1% is significant, * represents that the level of 10% is significant.

.

In Table 4, Column (1) contains regression results without adding control variables, and Columns (2) to (4) contain regression results with gradually adding control variables. The regression coefficients of Mp_it to MGL_it are still not significantly negative regardless of whether control variables are added. The benchmark regression results showed that the national marine pasture demonstration zone policy could not significantly promote the growth of the green level of mariculture, and Hypothesis 1 was confirmed. The regression results of the control variables showed that the use of fish drugs hindered the improvement of the green level of mariculture, which was closely related to the problems of the irrational use of fish drugs and guiding the healthy development of the fishery drugs industry. The intermediate consumption of mariculture and per capita disposable income of fishermen have insignificant positive effects on mariculture, which may be due to the high environmental cost caused by the increase in intermediate consumption and per capita income. Compared with the absolute increase in economic level, the green level of mariculture is more sensitive to pollutants and carbon emissions.

The reasons for the failure of the policy to significantly promote marine green aquaculture are inseparable from the immature market conditions for the green transformation of marine aquaculture in China, and they are inseparable from the initial development stage of China’s national marine pasture demonstration areas. First, a good reputation mechanism requires relevant government departments and media to build bridges between enterprises and the masses and do a good job in information media. With the rise of social attention, industries and enterprises will pay more attention to their reputation and external image and take more consideration of the benefits and risks brought by social supervision. However, at present, society is still not paying enough attention to marine pasture and mariculture, and the industry and enterprises are not sensitive to external public opinions and lack the initiative and enthusiasm for the green transformation of mariculture [60]. Second, an adequate incentive mechanism requires the active guidance of the government to set up a series of incentive indicators to provide enterprises with various incentive measures, including interest rate concessions, to stimulate the motivation of enterprises to actively transform. Also, the incentive for the green transformation of mariculture is weak in all regions, and the incentive intensity varies from place to place. Third, a complete regulatory mechanism requires the government to develop a series of regulatory systems and take regulatory measures to standardize industry and enterprise behavior. China’s marine green aquaculture supervision started late, lacking a clear definition of pollution discharge and control links, and it is prone to problems such as an unclear division of labor and weak regulatory awareness [61]. To sum up, the imperfect reputation, incentive, and supervision mechanisms fill the road of the green transformation of mariculture in China with obstacles and thorns. In addition, the current phenomenon of “heavy economic benefits, light ecological benefits” still exists in China’s marine pasture and mariculture, which further reflects the urgency and necessity of the green transformation of mariculture.

4.3. Parallel Trend Test and Dynamic Effect

In order to verify the applicability of multi-period DID, this paper uses the Event Study Approach proposed by Lalonde et al. [62] to conduct a parallel trend test. The construction model is as follows:

$$MG{L}_{it}={\beta }_0+{\displaystyle \sum _{-3\le k\le 3}{\beta }_{1}M{p}_{i,t-1}}+\theta Control{s}_{it}+{\omega }_i+{\upsilon }_t+{\epsilon }_{it}$$

(8)

In Equation (8), Mp_i,t−1 is the dummy variable of |k| years before and after the approval of establishing the national marine pasture demonstration area, and the meanings of the other symbols are the same as those in Equation (7). When the processing group of province i is in the policy implementation year (t − k), the value of Mp_it is 1, otherwise it is 0. In this paper, the effects of policies in the three years before and three years after the implementation are selected for analysis. The results of the parallel trend test are shown in Figure 3.

It can be seen that before the implementation of the policy, there is no significant difference in MGL_it between the treatment group and the control group, and the model passes the equilibrium trend test. In addition, this paper lists the results of dynamic regression, as shown in

Table 5. Dynamic effects.

	(1)	(2)	(3)	(4)
Variable	MGL	MGL	MGL	MGL
T + 0	0.020 (0.132)	0.006 (0.043)	−0.167 (−0.803)	−0.273 (−0.790)
T + 1	0.084 (0.550)	0.074 (0.496)	−0.087 (−0.415)	−0.188 (−0.541)
T + 2	0.078 (0.525)	0.071 (0.484)	−0.112 (−0.466)	−0.210 (−0.573)
T + 3	0.042 (0.403)	0.045 (0.423)	−0.167 (−0.610)	−0.281 (−0.692)
Controls	NO	YES	YES	YES
Province FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
N	126	126	126	126

.

The results in Table 5 show that after considering the time lag of policies, the national marine pasture demonstration zones still have no significant promoting effect on the green level of mariculture. The reasons are as follows: First, in terms of the length of the observation period, the first batch of national marine pasture demonstration zones in China were established at the end of 2015, and marine pasture is a systematic project with a long return cycle. Therefore, the national marine pasture demonstration area, which is in the initial stage of development, does not have enough of a foundation and enough results to promote marine green aquaculture, but also needs long-term and persistent construction and development. Second, from the perspective of policy formulation, the green transformation of mariculture in China is advancing in an orderly manner, but the policy formulation of marine pastures still focuses on the design and planning of the overall goal, and the corresponding supporting policies and institutional mechanisms are not perfect; especially the policies on ecological environment supervision and comprehensive management of key sea areas are not mature. Third, from the perspective of policy implementation, the efforts of most regions in marine green aquaculture still remain in the implementation of the policy instructions of various ministries and commissions and the formulation of local policies, the intensity of policy implementation is not strong, and the intensity of policy implementation varies across the region. By comparison, it can be found that the province with the largest number of establishment during the observation period is Shandong, which has set up a total of 44 national marine pasture demonstration zones, while Fujian and Hainan, which have set up the least number of national marine pasture demonstration zones, have only set up 1 national marine pasture demonstration zone, respectively. The regression results with control variables showed that the establishment of national marine pasture demonstration zones had no significant dynamic effect on the green level of mariculture. It indicates that in the early stage of development, the national marine pasture demonstration zone policy could not significantly promote the growth of the green level of mariculture. Therefore, on the one hand, long-term policy guidance is needed to gradually improve a series of market conditions such as reputation, incentive, and regulatory mechanisms. On the other hand, it is also necessary to unswervingly implement the national marine pasture demonstration zone policy, actively explore the marine pasture planning and construction mode, strengthen the exchange of experience and cooperation between marine pastures, and give full play to the “experimental field” and demonstration role of the policy.

4.4. Robustness Test

When the TWFE model is used to estimate the effect of multi-period DID treatment, it is difficult to satisfy the two conditions of the intertemporal homogeneity of treatment effect and the independence of the control variables, which leads to bias in the evaluation results. In order to alleviate the possible bias in TWFE estimation of the effect of multi-period DID processing, the multi-period DID double robust estimator proposed by Callaway and Sant’Anna [63] was used in this paper for robustness tests. Double robust estimation will divide the sample into different groups, estimate the average treatment effect ATT(g) of the different groups, respectively, and then sum the ATT(g) of the different groups through corresponding strategies, reduce the sum weight of the ATT(g) of those groups that may have bias, and calculate the ATT of the sample period. Therefore, compared with the TWFE model, dual robustness estimation can obtain more efficient and consistent estimators. As can be seen from

Table 6. Dual robust estimators of multi-period DID.

	(1)	(2)
Variable	MGL	MGL
ATT	−0.052 (−0.861)	−0.075 (−1.113)
Controls	NO	YES
N	126	126

, no matter whether control variables are added or not, the average treatment effect is not significantly negative, and the national marine pasture demonstration area still does not significantly promote the growth of the green level of mariculture, which further supports the above analysis results.

4.5. Mechanism Test

On the basis of differentials, the construction of a triple difference model can further aid in the study of the mechanism of policy [64], so as to better evaluate the effect of policy. Therefore, based on the treatment method of Cai et al. [65] and the triple difference model, this paper examines the potential mechanism of the influence of policies on marine green aquaculture by constructing the cross-terms between the establishment of national marine pasture demonstration zones and the institutional environment, industrial structure, and the proxy variables of scientific and technological innovation, respectively. The selection of proxy variables of the mechanism is explained as follows: the ratio of the aquaculture technology extension funds to the aquaculture area is used to characterize the institutional environment. As a financial support budget, aquaculture technology extension funds are not only for the consideration of the comprehensive benefits of mariculture, but also for the government’s active actions in improving the institutional mechanism, which better reflects the status of the local institutional environment. In order to reflect the contribution of mariculture in regional production, the ratio of the total output value of mariculture to the gross regional product was used to characterize the industrial structure. Patents can fully reflect the level of scientific and technological innovation of marine pastures, so the number of scientific and technological patents granted by marine scientific research institutions is used to represent scientific and technological innovation. The three proxy variables are all logized, as shown in Table 3. To sum up, the mechanism test model is as follows:

$$MG{L}_{it}={\beta }_0+{\beta }_{1}M{p}_{it}+{\beta }_{2}M{p}_{it}\ast Mec+{\beta }_{3}Mec+\theta Control{s}_{it}+{\omega }_i+{\upsilon }_t+{\epsilon }_{it}$$

(9)

In Equation (9), Mec represents the proxy variable of the three mechanisms, and the other symbols have the same meaning as in Equation (7). Mp_it × FTP, Mp_it × TIS, and Mp_it × STO represent the institutional environment, industrial structure, and scientific and technological innovation, respectively. As can be seen from

Table 7. Mechanism test.

	(1)	(2)	(3)
Variable	MGL	MGL	MGL
Mpit × FTP	−0.273 (−0.005)
Mpit × TIS		16.336 (1.265)
Mpit × STO			0.086 (1.696)
Constant	−1.396 (−0.393)	−2.349 (−0.699)	−2.497 (−0.782)
Controls	YES	YES	YES
Province FE	YES	YES	YES
Year FE	YES	YES	YES
N	126	126	126
R2	0.059	0.106	0.163

, the regression coefficients of the influence effects are not significant, which indicates that the policy of the national marine pasture demonstration area has not significantly promoted the growth of the green level of mariculture under the influence of the institutional environment, industrial structure, and scientific and technological innovation. Hypothesis 2 is confirmed.

In order to realize the win–win situation of ecological environment and fishery resources, the construction of marine pasture is an international practice. In addition to increasing marine fisheries resources, marine pasture has two core purposes: ecological restoration and the conservation of biological resources. Especially in the process of practicing the “two mountains theory” and “double carbon target”, the concept of marine pasture and marine green aquaculture is highly consistent, and the construction of marine pasture should provide a solid driving force for marine green aquaculture. The reasons for the policy of the national marine pasture demonstration zone affecting the potential mechanism failure of marine green aquaculture are as follows:

From the perspective of the institutional environment, China’s marine pasture started late. Due to the shortage of funds and lack of experience in the 1990s, the construction of marine pasture once stalled, and it was not until this century that the construction boom was set off. In some areas, there is a phenomenon of “heavy construction and short-term economic interests, light management and long-term overall planning”, which has diluted the comprehensive benefits brought by marine pastures to promote “green farming” to a certain extent. In the process of the rapid development of marine pastures, the defects of the imperfect system and mechanism gradually appear. Only with the joint cooperation of relevant stakeholders such as the government, fishing companies, and fishermen can marine ranches avoid the construction and operation of separate policies. When there is a certain lack of management and overall planning, it is difficult to ensure the scientific demonstration of planning layout, construction scale, and population selection. There is therefore a need to create an enabling institutional environment to co-ordinate and manage marine pastures. However, the construction of marine pasture laws and regulations is lagging behind, which can easily produce problems such as an unclear division of rights and responsibilities, chaotic management, and insufficient supervision. The excessively high standard of royalty and the excessively short period of allowable use of the sea area are both manifestations of the lack of laws and regulations on marine pasture [47]. It can be seen that the imperfect institutional environment is not conducive to the policy of national marine pasture demonstration areas to promote marine green aquaculture.

From the perspective of industrial structure, marine pasture is of great help to the transformation and upgrading of fishery industrial structure. With the deepening of the construction of different types of marine pastures, intelligent farming, fishing transformation, intensive processing, and fishery leisure have been strongly promoted, effectively adjusting the industrial structure of marine fisheries. According to the function standard, marine pasture can be divided into fishery cultivation, ecological restoration, germplasm protection, leisure, and sightseeing. In order to match the functions of the ranch, the marine environment, and the development goals, different types of equipment and facilities need to be equipped, which puts higher requirements on the intelligent equipment and supporting facilities of the marine ranch. However, the foundation of intelligent equipment for marine pastures in China is weak, and there is a certain gap from “digitalization and systematization”. In recent years, with the investment of a large amount of funds, the equipment and facilities of marine ranches have been greatly improved. However, the high development cost and the excessive number of artificial fish reefs are the main reasons for the inefficiency of marine pastures. In the early stage of development, intelligence may be locked in primary technologies, models, and paths by blind and excessive investment, increasing the difficulty of future updates [66]. Once trapped in the low-end lock of intelligence and equipment facilities, ocean ranches will lack sufficient driving forces, and then there will be diminishing marginal returns. In the early stage of the development of marine pastures, the weak foundation and blind construction blocked the effect of industrial structure to promote marine green aquaculture.

In terms of technological innovation, early sea ranching was seen as an ocean system that could increase the amount of food available. As people gradually realize the importance of science and technology in marine pastures, marine pastures need science and technology to achieve ecological health, resource abundance, food safety, and other goals [49]. Whether it is Ocean Ranch 2.0, which emphasizes ecology and informatization, or Ocean Ranch 3.0, which emphasizes digitalization and systematization, all show the importance of the integration of technology and ecology in the development of the ocean ranch. However, the scientific and technological foundation of China’s marine ranches is weak. According to the CNKI patent database, there were only 6 marine pasture patents in the country in 2012, but the development speed was fast, the number of patents reached 50 in 2018, and the number of patents exceeded 100 for the first time in 2021. At the same time, China’s marine pastures are developing rapidly, and the construction funds and the number of reefs are increasing year by year. By 2020, all coastal provinces in China had built national-level marine pasture demonstration zones. Although science and technology have made great progress, scientific and technological support still encounter difficulty in meeting the needs of the rapid development of marine pastures. The key basic technologies such as fishery tracking and control, submarine structure, fishing forecasting, and early warning have not provided breakthroughs, and the scientific and technological shortcomings have seriously restricted the effect of marine pasture in promoting marine green aquaculture.

4.6. Heterogeneity Analysis of Marine Pastures in the North and South

In this paper, the national marine pasture demonstration areas in the north and south are grouped to observe the policy effect. Mp_it_north and Mp_it_south are the core estimators using the TWFE model, ATT_north and ATT_south are the average treatment effects of the dual robust estimators, and north and south represent the national marine pasture demonstration zones in the north and south, respectively. The specific regression results are shown in

Table 8. Heterogeneity test.

	(1)	(2)	(3)	(4)
Variable	MGL	MGL	MGL	MGL
Mpit_north	−0.161 (−0.910)	−0.125 (−0.822)
ATT_north	−0.192 * (−1.868)	−0.219 * (−1.886)
Mpit_south			0.120 (0.986)	0.081 (0.813)
ATT_south			0.113 (1.639)	1.514 * (1.728)
Constant	1.355 *** (18.316)	−1.812 (−0.483)	1.234 *** (16.225)	−1.913 (−0.507)
Controls	NO	YES	NO	YES
Province FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
N	126	126	126	126
R2	0.058	0.082	0.046	0.071

Notes: The t value is in parentheses. *** represents that the level of 1% is significant, * represents that the level of 10% is significant.

.

It can be seen from Table 8 that no matter whether TWFE model or double robust estimator is used, the regression coefficients of the northern national marine pasture demonstration area are negative, while those of the southern national marine pasture demonstration area are positive, and the significance of the regression coefficients of the double robust estimator is significantly higher than that of the TWFE model. The regression coefficients of the dual robust estimators were observed, and it was found that the regression coefficients of the northern national marine pasture demonstration area were significantly negative regardless of whether the control variables were added. For the southern national marine pasture demonstration area, the regression coefficient was significantly positive after adding control variables. This indicates that the national marine pasture demonstration zone in northern China has not significantly promoted the growth of the green level of mariculture, but has a certain hindering effect and falls into a “policy trap”. The southern national marine pasture demonstration zone has significantly promoted marine green aquaculture. In summary, Hypothesis 3 is proved.

There are three possible reasons for the northern national marine pasture demonstration zone to fall into the “policy trap”: first, the northern marine pasture pays more attention to the proliferation function and is mostly small and medium-sized ecological restoration marine pasture, with a poor scale effect. In addition, marine pastures are mostly distributed near the sea, which is easily affected by land-based pollution. Second, there is the problem of “heavy construction, light management”, blindly putting artificial reefs into the early stage of construction, which not only violates intensive management, but also causes the unreasonable development and utilization of marine pastures. Third, if the early investment is unreasonable, it may cause pressure on the subsequent construction progress, affect the update of technology and equipment facilities, and produce low-end locking problems. There are three possible reasons for the southern national marine pasture demonstration zone to help marine green aquaculture: First, the southern marine pasture pays more attention to ecological conservation, there are a large number of conservation marine pastures, and the large and medium-sized ecological restoration marine pastures are mainly offshore, so it is easier to realize the scale effect of green farming and ecological conservation. Second, due to the unique geographical conditions, the southern ocean ranch will pay more attention to the rational development and utilization of sea areas and islands in planning and construction, and they will pay more attention to the intensive cultivation mode. Third, the southern coastline is more tortuous, and large marine pastures are mainly distributed in the open sea, which is not easily affected by land source pollution, which provides necessary conditions for green transformation and ecological conservation.

5. Research Conclusions and Policy Implications

5.1. Research Conclusions

The main conclusions are as follows:

(1)

The green level of mariculture in our country is of low efficiency. The empirical results of multiple periods of DID showed that the establishment of national marine pasture demonstration areas did not significantly promote the growth of the mariculture level, and the hypothesis of the influence mechanism was verified by analyzing the evolutionary game model. Parallel trend test, dynamic effect test, and multi-period DID double robust estimation were used to test robustness. The conclusions obtained were consistent with the baseline regression results, which verified the robustness of the baseline regression results.

(2)

A poor institutional environment, unreasonable industrial structure, and insufficient scientific and technological innovation are the main reasons that lead to the failure of national marine pasture demonstration zones to significantly promote marine green aquaculture. From the perspective of the institutional environment, management and overall planning are insufficient, and the construction of laws and regulations lags behind, resulting in a lack of institutional mechanisms to promote marine green aquaculture. From the perspective of industrial structure, the weak foundation, blind construction, and high cost block the effect of industrial structure to promote marine green aquaculture. From the perspective of scientific and technological innovation, the scientific and technological foundation of marine ranches is weak, the key technologies have not broken through, and the short board of science and technology has difficulty in meeting the needs of the rapid development of marine ranches.

(3)

The results of heterogeneity analysis showed that there were differences in the effect of policies on the green level of mariculture in the northern and southern national marine pasture demonstration zones. Affected by the differences in planning objectives, resource endowments, location conditions, and operation scale, the northern national marine pasture demonstration zone did not significantly promote marine green aquaculture, while the southern national marine pasture demonstration zone significantly promoted marine green aquaculture.

5.2. Policy Implications

According to the research conclusions, the following three policy implications can be obtained:

(1)

Create a good market environment for the green transformation of mariculture and give full play to the promotion role of the national marine pasture demonstration zone policy. First, pay attention to the reputation mechanism. First of all, the government should play a leading and regulating role to ensure that the public and the news media actively participate in social supervision. Publicity and education should be strengthened to increase public attention and participation in marine ranches through popular science. We should improve the complaint reporting system, establishing online and offline multi-channel supervision and reporting channels. Second, we should improve the incentive mechanism. The majority of implementing subjects should put the incentive mechanism into practice, accurately convey the relevant system to mariculture enterprises and mariculture farmers, and fully mobilize the enthusiasm of participation. Third, we should innovate the regulatory mechanism. We will accelerate the establishment of a monitoring system for multi-period analysis, dynamically monitor the operation of multi-period analysis, and ensure the scientific and accurate evaluation of multi-period analysis benefits.

(2)

Improve systems and mechanisms to turn disadvantages into advantages. First, we should build a complete system of laws, regulations and supervision. First of all, the management methods of marine pasture construction should be written into the high-level law, and the introduction and implementation of supporting laws and regulations should be promoted to clarify the rights and responsibilities of all parties. Secondly, the strategic layout of marine pastures within and outside the region should be coordinated to change the phenomenon of local “light management and overall planning”. Finally, we should strengthen the evaluation of the applicability of existing management methods and regulatory systems and make plans for the long-term improvement of legislation. Second, we should promote the construction of intelligent marine pastures and accelerate the transformation and upgrading of the industrial structure. While increasing investment in intelligent equipment and supporting facilities for marine ranches, we will actively explore new models of integrated development such as offshore wind power, interaction between breeding and grazing, and driving by seed industries. Third, we should adhere to innovation-driven development and attach importance to technological system innovation. On the one hand, sufficient research and development funds should be provided to help overcome the key technologies of marine pasture. On the other hand, we should not only pay attention to micro technological innovation, but also promote the innovation of macro technological systems.

(3)

Combined with resource endowments, guide the green transformation of mariculture according to local conditions. First, northern marine rangelands should strengthen ecological conservation and improve regulatory efficiency. Firstly, the spatial planning of multi-period analysis should be optimized to promote the formation of scale effects of ecological conservation and green aquaculture. Second, we should promote the development and application of new energy, new materials, new technologies, and new equipment to help achieve “win–win” economic and ecological benefits. Finally, the concept of “heavy construction and light management” should be changed to strengthen the scientific control of early input costs and realize the supervision of the whole process of marine ranch construction. Second, southern marine ranches should give full play to the advantages of ecological conservation and be on guard against falling into policy traps due to improper resource development and utilization. On the one hand, we should continue to maintain the good momentum of the development of large and medium-sized ecological restoration marine pastures, make good use of the advantages of industrial capital and business environment, and promote the research and development of core technologies and the transformation of achievements. On the other hand, we should make full use of sea space and resources, pay attention to the rational development and ecological conservation of islands and reefs, and strengthen intensive management.