Evidence of the Contribution of the Technological Progress on Aquaculture Production for Economic Development in China—Research Based on the Transcendental Logarithmic Production Function Method

Abstract

Technology includes hard technology and soft technology. Material technology embodied in production conditions and working conditions such as machinery, equipment and infrastructure is called hard technology, referring to the technology directly used in the development and production of means of production and means of subsistence, such as product design technology, equipment manufacturing technology, etc. Non-material technology, which embodies the experience, skills and management ability of process management, decision support and information technology, is called soft technology. Hard technologies make things easier and faster, while soft technologies promote flexibility and creativity. However, hard technologies take time to produce and have negative environmental impacts, while soft technologies are simple to produce but hard to use. Hence, finding the right balance between hard and soft technology investment is important for the sustainable increase of productivity. In recent years, China has continuously increased its investment in science and technology in aquaculture industry. However, the majority of the investment has gone to hard technology, which has hampered the long-term development of the industry. This paper aims to look at the status quo of the scientific and technological progress of the aquaculture industry in China and explore how the advancement of hard and soft technologies play a role in the economic growth of the aquaculture industry. Transcendental logarithmic production function is employed to calculate the contribution rate of technological progress on China’s aquaculture industry, and the contribution rates of hard technologies and soft technologies are examined separately. The results indicate that, from 2012 to 2020, the contribution rate of overall technical progress on aquaculture in China was 80.159%, of which 71.720% came from the progress of hard technologies, while only 8.439% came from the progress of soft technologies. Based on this conclusion, the paper calls for a balance of hard and soft technologies in the aquaculture industry in China to ensure a healthy and sustainable aquaculture industry. Policy suggestions are also put forward.

1. Introduction

Aquaculture is an important source of aquatic products in China. In 2021, the output of cultured aquatic products reached 53,949,000 tons, which was 4.2 times the output of fishing, accounting for 80.63% of the total output of aquatic products in China. Aquaculture plays an increasingly important role in the production and supply of aquatic products in China. To establish the concept of “big food”, it is not just the supply of grain, but also the effective supply of aquatic products and other foods, that should be ensured. Due to the rising global food price in recent years and the increasingly reliance on imports for feed grain, aquaculture has become a veritable high-efficient form of agriculture because aquaculture produces higher-quality protein with less resource. In fact, aquaculture has been seen as a great contributor to the optimization of agricultural industrial structure, the promotion of rural economic construction, the development of foreign trade and the increases in farmers’ incomes [1].

In the past decade, the aquaculture area, the aquaculture population and other input in aquaculture in China have been decreasing year by year (see

Table 1. Factor input data of aquaculture in China from 2012 to 2020.

Year	Aquaculture Area (ha)	Aquaculture Population	Fish Species Input
2012	8,088,403	5,214,333	112,302,142.1
2013	8,321,699	5,191,739	191,998,186.7
2014	8,386,360	5,124,211	128,109,780.5
2015	8,465,004	5,103,175	127,436,814
2016	8,346,339	5,021,686	130,266,310.7
2017	7,431,630	4,901,871	133,185,902.9
2018	7,189,524	4,742,727	132,382,397
2019	7,108,497	4,663,678	126,313,059.7
2020	7,036,106	4,575,402	132,135,566
Average annual growth rate%	−1.73	−1.62	2.05

Source: China Fishery Statistics Yearbook [2].

). The aquaculture area is diminishing, and the economic growth of aquaculture is dependent (to a greater extent) on the improvement of science and technology. According to the law of diminishing marginal returns, the marginal value gradually decreases if the input of factors continuously increases. Coupled with the limited land and water resources present in China, the increasing costs of labor and the environmental cost, the economic growth of aquaculture in China can only be achieved by ensuring technological progress in aquaculture production. The contribution rate of technological progress is an important index that reflects the effect of scientific and technological progress on the promotion of industrial development. Technological progress can be divided into hard technological progress (or narrow technological progress) and soft technological progress. China has continuously increased its investment in fishery science and technology, strengthened technological innovation in fishery and comprehensively improved the product quality, efficiency and competitiveness of its fishery sector. According to the data obtained from the China Fishery Statistics Yearbook, in 2021, the financial allocation was 1.771 billion yuan for the construction of demonstration bases and laboratories and the promotion of demonstration key technologies. However, what is the status quo of the scientific and technological progress of the aquaculture industry in China? How does the advancement of technologies (i.e., hard technology and soft technology) play a role in the economic growth of the aquaculture industry? These are the questions that this paper aims to answer.

2. Theoretical Analysis and Literature Review

2.1. Division of Technological Progress

Technology refers to the process that transforms inputs into outputs. Technological progress refers to the discovery of new and advanced methods of producing goods. The improvement in technology leads to an increase in productivity of capital, labor and other production factors. According to the classification method of broad science and technology [3], science and technology have been divided into soft science and technology and narrow science and technology (or hard science and technology). Some differentiate hard and soft science and technology based on whether or not the science and technology have the shape of an object (the ones with the shape of an object are called hard science and technology, and those without the shape of an object are called soft science and technology). For example, Yin [4] thinks that hard technology is represented by materials, including the innovation of core products and manufacturing equipment, while the progress of soft technology is manifested in the process of production, management, promotion and even marketing. Zhong and Jiang [5] consider promoting soft technological progress is equivalent to improving management skills. Guo [6] believes that soft technology uses new scientific energy-saving management methods to improve energy-saving efficiency, which is significantly different from hard technologies such as new methods, new processes, new equipment manufacturing and new product development. Sun [7] defines R&D and the application of new technologies as hard technological progress, and regards the improvement of technical efficiency, the efficiency of resource allocation, the efficiency of product structure and the rate of return on scale as soft technological progress.

Based on the above literature, this paper divides the broad fishery technological progress into fishery hard technological progress and fishery soft technological progress. Some examples of hard technologies are new aquaculture technology, new equipment manufacturing and new product development, etc. (there is an emphasis on invention and transformation in this definition). Some examples of soft technologies are management, aquatic science and technology extension service, information science and technology decision support, etc. Together, they constitute the technical progress of fishery in a broad sense.

In the past decade, the progress of hard and soft technologies in aquaculture in China has included, but is not limited to: (1) Innovating the high-density ecologically healthy aquaculture mode such as industrial circulating water aquaculture, improving effective utilization of land resources and water resources and realizing high-density and intensive aquaculture; (2) innovating the rapid detection technology of aquatic animal diseases, greatly improving China’s ability to deal with aquatic animal diseases quickly, effectively avoiding the outbreak of aquatic diseases and thereby reducing financial losses caused by the diseases; (3) micro-hole oxygenation technology, which has improved the quantity of dissolved oxygen at the bottom of aquaculture water, improved the quality of aquaculture water, further increased the breeding yield and reduced the energy consumption of aquaculture; (4) internet of things (IoT) and other information technologies that are widely used in aquaculture production activities, as real-time monitoring of aquaculture environment and the intelligent management of aquaculture process have become a reality and resulted in aquaculture quality being improved and aquaculture risks being greatly reduced; (5) establishing a fishery technology extension service network in the field of fishery organization and management, led by the central and local government and allowing the role of demonstration and extension of new technologies to be actively improved. Production entities are increasingly diversified, and new agricultural business entities, such as large professional households, family farms, farmers’ professional cooperatives and agricultural enterprises, are constantly emerging. A new industrial organization model (i.e., agricultural industrialization consortium) has been established, which has effectively promoted the high-quality development of fisheries.

2.2. Measurement and Decomposition of Contribution of Technological Progress

The rate of technological progress was first put forward by Thoreau [8] in 1957. This term describes the contribution of scientific and technological progress to economic development. The ratio of the technological progress growth rate to the economic growth rate is the contribution rate of technological progress [9,10]. Scholars at home and abroad have studied the rate of technological progress and its contribution. Foreign scholars have made pioneering contributions mainly in the aspects of method innovation and the analysis of technological progress factors. Kunimitsu [11] included climate factors in his research of Japanese rice models, which has been regarded as a more accurate measurement of the contribution rate of scientific and technological progress. In 1986, Griliches [12] empirically studied manufacturing survey data in the United States and found that scientific and technological progress had a significant impact on improving productivity. Kumbhakar et al. [13] converts total factor productivity into technological progress, technological efficiency, scale efficiency and allocation efficiency, which provides a method to calculate technological progress. Vinay [14] assessed the rate of technological progress in the tuna fishery in the Laksha Islands and analyzed the reasons for its economic growth as a result of technological developments in the areas of introduction of new mechanical sprayers, the improvement of engines, the development of infrastructure (such as cold storage and ice plants) and the recognition of the potential of yellowfin tuna resources. The research on the contribution rate of China’s technological progress started later, mainly by learning and introducing the theories and methods of foreign scholars. For example, Meng Lingjie [15] and Yang Xuejiao [16] used the data envelopment analysis (DEA) method to analyze the change and influence of the contribution rate of agricultural technological progress in China; meanwhile, Gao Yang [17] and Guo Jiang [18] employed the Solow residual method to calculate the contribution rate of agricultural and rural scientific and technological progress in Shanxi and Shaanxi. On the decomposition of the contribution rate of technological progress, Zhao Zhijun [19] calculated the contribution rate of rural science and technology in China from 1985 to 2005, and divided it into the narrow contribution rate of scientific and technological progress, the rate of change of scientific and technological benefits and the rate of change of return on scale. Su Feifei [20] combined the quality of labor force and national policy as modifying variables with agricultural employees and agricultural material consumption, and found that the contribution rate of agricultural technological progress declined somewhat; by using the improved multi-factor second-order constant elasticity of substitution (CES) production function, Zhang Ziqiang [21] calculated the output elasticity of forestry input factors and the contribution rate of forestry technological progress. Mu Zongzhao [22] found the contribution of forestry science and technological progress of the “Forest Resources Development and Protection Project” in Shandong Province to be 41.39% by employing the growth rate equation.

In the fishery field, Xiang Wenqi [23] defined the contribution of scientific and technological progress to fishery economic growth as the contribution rate of fishery technological progress. Generally, the research on the contribution rate of fishery technological progress focuses mainly on the contribution of broad technological progress. For example, Zhou Qicai [24] used the Solo residual method to calculate the contribution rate of fishery technological progress and aquaculture and concluded that the contribution of fishery technological progress to fishery is higher than that to aquaculture. Yang Wei and Xiang Wenqi [25] used the Cobb-Douglas (C-D) production function method to calculate the contribution rate of scientific and technological progress of China’s marine fisheries, which increased sharply from -93.65% during the Tenth Five-Year Plan period to 4.18% during the Eleventh Five-Year Plan period, indicating that the scientific and technological contribution of China’s marine fisheries was appearing gradually, but it was still at a very low level. Zhu Yuchun [26] estimated that the contribution rate of technological progress of freshwater aquaculture in China from 1990 to 2007 was 60.87%. Liu Zifei [27] compared the previous calculation methods of the contribution rate of scientific and technological progress, and proposed to calculate the contribution rate of fishery scientific and technological progress by region, operation type and time period to ensure the accuracy of the calculation results.

2.3. Significance of This Research

The existing literature has explored the contribution rate of the technological progress of planting, forestry and fishery, but there is a lack of analysis on the contribution rate of technological progress of aquaculture. At present, China’s fisheries include fishing and aquaculture. In China, the fishing industry adopts the summer fishing moratorium system and limited fishing to protect fishery resources, so the income of fishermen depends more on aquaculture based on the reality of fishery development in China. This paper analyzes the contribution rate of technological progress of aquaculture in China, which has specific, practical significance for improving the income of fishermen.

The main contributions of this paper are as follows: 1. Systematically analyzing the contribution of science and technology input to aquaculture in China in recent decades; 2. Differentiating the contribution rates of hard technological progress and soft technological progress, and discussing the effects of both types of technological progress on aquaculture’s economic growth; and 3. Since most literature has regarded the number of employees as the input of labor force, but ignored the importance of labor quality for the fishery production process, taking the combination of the average years of education of labor force in each province and the number of fishery breeding population in each province as a new input variable of labor force based on the new economic growth theory, which makes the calculated contribution rate of technological progress closer to reality.

3. Research Methods and Data Sources

3.1. Model Setting of Contribution Rate of Fishery Technological Progress

There are two methods used to measure the rate of technological progress: the nonparametric method and the parametric method. For nonparametric methods, the DEA method and the Malmquist exponent method are mainly used. Nonparametric methods cannot test the applicability of the frontier and do not consider the influence of random factors on the measurement results. Their advantage is objectivity and their disadvantage is inflexibility. For parametric methods, the Solow residual method, the recessive variable method and the stochastic frontier production function method are mainly used. The Solow Residual Method is based on the assumption that the market is completely competitive and the technological progress is Hicks neutral and the return to scale is constant. The residual obtained from the C-D production function is regarded as total factor productivity. These strict assumptions are often inconsistent with reality. The fact that the total factor productivity is obtained when both labor and capital are fully utilized is only explained as the influence of technological progress. However, total factor productivity includes not only technological progress but also economies of scale and efficiency improvement. The Solow residual method does not eliminate the influence of calculation errors on total factor productivity, which will lead to overestimation of technological progress. The stochastic frontier approach (SFA) sets the form of production function as transcending logarithmic production function, which is more flexible than the C-D function, relaxes the assumptions of constant returns to scale and technology neutrality and allows for underutilization of labor and capital.

This paper uses the transcendental logarithm (Translog PF) production function to break down the generalized technological progress into hard technological progress and soft technological progress and to analyze the contribution of both types of technological progresses to the economic growth of aquaculture. The method was first put forward by L. Christensen, D. Jorgenson and Lau [28]. The transcendental logarithmic production function is mainly constructed to reflect the elasticity of factors, the temporal trend of technological progress in a narrow sense and the interaction among factors, and it is a variable elastic production function model that is easy to estimate and has strong inclusiveness. The function form is as follows (Formula (1)). Its specific technical types were not strictly defined in the function modeling, and it was developed according to the second-order Taylor formula, in which the expansion term included not only the interaction between technological progress terms and elements but also Hicks neutral technological progress terms [29]. The model form of this method is defined as follows:

Let the abstract production function be Y = F (lnK, lnL, lnN, lnA,), where A represents the technical level changing with time. Let the logarithms on both sides of the equation be taken, that is, LnY = LnF (lnK, lnL, lnN, lnA). According to the second-order Taylor formula, the final expansion is obtained:

LnY = α₀ + α₁t + 1/2α₂t² + β_KlnK + β_AKtlnk + β_Ltlnk + β_ALtlnL +
β_NlnN + β_ANlnN + 1/2β_KKln²k + 1/2β_LLtln²L + 1/2β_NNln²N + β_KLlnKlnL +
β_KNlnKlnN + β_LNlnLlnN + ηdisa + v − μ

(1)

where Y is the output value of aquaculture and t is a time variable, indicating the change of technological progress with time. N represents land input, K represents material input, L represents labor factor input and β is the parameter to be estimated. Among them, β_KL, β_KN and β_LN represent the interaction between capital, land and labor factors put into production. The interaction effects between the invested capital, labor and land factors and the technical level are expressed by β_AK, β_AL and β_AN, respectively, and the self-strengthening of production factors is expressed by β_KK, β_LL and β_NN, respectively [18].

By taking the derivative of t in Formula (1), the calculation formula of the hard technological progress rate coefficient can be obtained by interpolation:

$$\mathrm{TP}=\frac{\partial \mathrm{lnY}}{\partial \mathrm{t}}=({\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1\mathrm{t})+({\mathsf{\beta }}_{\mathrm{AK}}\frac{\mathrm{dlnK}}{\mathrm{dt}}+{\mathsf{\beta }}_{\mathrm{AL}}\frac{\mathrm{dlnL}}{\mathrm{dt}}+{\mathsf{\beta }}_{\mathrm{AN}}\frac{\mathrm{dlnN}}{\mathrm{dt}})$$

(2)

where $\left({\mathsf{\alpha }}_1+{\mathsf{\alpha }}_2\mathrm{t}\right)$ it represents the neutral technological progress that changes with time, (β_AK$\frac{\mathrm{dlnK}}{\mathrm{dt}}$ + β_AL$\frac{\mathrm{dlnL}}{\mathrm{dt}}$ + β_AN$\frac{\mathrm{dlnN}}{\mathrm{dt}}$) it represents the total partial effect of scientific and technological progress.

According to the above two formulas, the contribution rate of fishery hard technological progress is as follows (TPC stands for technological progress contribution rate).

$$\mathrm{TPC}=\frac{\mathrm{TP}}{\raisebox{1ex}{$\Delta \mathrm{Y}$}\!\left/ \!\raisebox{-1ex}{$\mathrm{Y}$}\right.}\times 100\%$$

(3)

The Divisia total factor productivity (TFP) index is introduced, and the generalized technological progress function is decomposed completely

$$\mathrm{TFP}=\frac{\mathrm{dlnY}}{\mathrm{dt}}-{{\displaystyle \sum }}_{\mathrm{i}}{\mathrm{s}}_{\mathrm{i}}\frac{{\mathrm{dlnX}}_{\mathrm{i}}}{\mathrm{dt}}$$

(4)

where ${\mathrm{s}}_{\mathrm{i}}=\frac{{\mathrm{w}}_{\mathrm{i}}{\mathrm{X}}_{\mathrm{i}}}{\sum {\mathrm{w}}_{\mathrm{i}}\mathrm{X}}$, ${\mathrm{w}}_{\mathrm{i}}$ is the price of factor i.

In the process of decomposing generalized technological progress, it is necessary to calculate the factor of technical efficiency. Thus, the stochastic frontier model is introduced, and its error term is divided into two parts: the technical incompletely controllable factor (V) and the technical efficiency (U) factor [30].

$$\mathrm{Y}=\mathrm{F}\left(\mathrm{X},\mathrm{t}; \mathsf{\beta }\right){\mathrm{e}}^{\mathrm{v}-\mathrm{u}}$$

After taking logarithms on both sides, its functional form is as follows:

lnF(X,t,β) + v − u where u > 0, t = 1,2, …,t

(5)

Next, calculate the derivative of Formula (5) with respect to time t. The formula for calculating the rate of change of technical efficiency is as follows:

$$\begin{array}{ll}\frac{\mathrm{dlnY}}{\mathrm{dt}}& =\frac{\partial \mathrm{lnF}}{\partial \mathrm{t}}+{\displaystyle \sum }_{\mathrm{i}}^{\mathrm{n}}\frac{\partial \mathrm{lnF}}{\partial {\mathrm{X}}_{\mathrm{i}}}\frac{{\mathrm{dlnX}}_{\mathrm{i}}}{\mathrm{dt}}-\frac{\partial \mathrm{u}}{\partial \mathrm{t}}\\ & =\mathrm{TP}+{\displaystyle \sum }_{\mathrm{i}}^{\mathrm{n}}{\mathsf{\epsilon }}_{\mathrm{i}}\frac{{\mathrm{dLnX}}_{\mathrm{i}}}{\mathrm{dt}}+\mathrm{TE}\end{array}$$

Bringing the Divisia total factor growth rate index into Equation (5) results in the following:

$$\mathrm{TFP}=\mathrm{TP}+{\displaystyle \sum }_{\mathrm{i}}^{\mathrm{n}}\left({\mathsf{\epsilon }}_{\mathrm{i}}-{\mathrm{s}}_{\mathrm{i}}\right)\frac{{\mathrm{dLnX}}_{\mathrm{i}}}{\mathrm{dt}}+\mathrm{TE}=\mathrm{TP}+\left(\mathsf{\epsilon }-1\right){\displaystyle \sum }_{\mathrm{i}}^{\mathrm{n}}\frac{\mathsf{\epsilon }\mathrm{i}}{\mathsf{\epsilon }}\frac{{\mathrm{dLnX}}_{\mathrm{i}}}{\mathrm{dt}}+{\displaystyle \sum }_{\mathrm{i}}^{\mathrm{n}}\left(\frac{\mathsf{\epsilon }\mathrm{i}}{\mathsf{\epsilon }}-{\mathrm{s}}_{\mathrm{i}}\right)\frac{{\mathrm{dLnX}}_{\mathrm{i}}}{\mathrm{dt}}+\mathrm{TE}$$

(6)

Among them, $\mathsf{\epsilon }=\sum {\mathsf{\epsilon }}_{\mathrm{i}}$ is the return to scale index. The progress rate of hard technology is TP and the change rate of technical efficiency is TE. The $\left(\mathsf{\epsilon }-1\right){{\displaystyle \sum }}_{\mathrm{i}}^{\mathrm{n}}\frac{\mathsf{\epsilon }\mathrm{i}}{\mathsf{\epsilon }}\frac{{\mathrm{dLnX}}_{\mathrm{i}}}{\mathrm{dt}}$ is the variable rate of return on scale. The ${{\displaystyle \sum }}_{\mathrm{i}}^{\mathrm{n}}\left(\frac{\mathsf{\epsilon }\mathrm{i}}{\mathsf{\epsilon }}-{\mathrm{s}}_{\mathrm{i}}\right)\frac{{\mathrm{dLnX}}_{\mathrm{i}}}{\mathrm{dt}}$ in the equation is the change of allocation efficiency of production factors.For the change rate of the allocation efficiency of production factors, Schultz’s rational small-scale peasant hypothesis theory holds that [31]: Under the long-term and certain limited environmental conditions encountered by farmers’ production, farmers’ production can always arrange all input factors in the equilibrium state of optimal prediction or achieve a good benefit state. Therefore, the annual average change rate of distribution efficiency of production factors fluctuates very little and can be ignored.

According to Kumbhakar’s decomposition formula [13], there are the following simplified formulas:

$$\mathrm{TFP}=\mathrm{TP}+\left(\mathsf{\epsilon }-1\right){\displaystyle \sum }_{\mathrm{i}}^{\mathrm{n}}\frac{\mathsf{\epsilon }\mathrm{i}}{\mathsf{\epsilon }}\frac{{\mathrm{dLnX}}_{\mathrm{i}}}{\mathrm{dt}}+\mathrm{TE}$$

(7)

This means that the sum of the first three data can represent the generalized rate of technological progress. According to Zhu Xigang’s [3] division of technological progress, we can see that the rate of hard technological progress is expressed by the first term TP on the right side of Equation (6); that is, the rate of hard technological progress. The second and third items are the content of the rate of soft technological progress.

Since there are two kinds of uncontrollable factors in the stochastic technology frontier model (namely, random error (V) and technical efficiency error (U)), the classical OLS method cannot make a reasonable estimation, so the maximum likelihood function estimation (MLE) method is used in this paper.

3.2. Variable Setting of Contribution Rate of Fishery Technological Progress

Lucas and Paul Romer’s [32] new economic growth theory holds that technological progress is the source of economic growth. They also expanded the definition of labor force to human capital investment; that is, labor force includes not only the absolute labor force quantity index, but also the education level, the production skill training and other quality indexes of labor force. In the traditional research on the contribution rate of fishery technological progress, the variables are generally capital investment, labor force quantity and land investment. The quality of the labor force is also a very important factor for fishery production. The quality of fishery personnel is the characteristic of each labor force, which directly affects the innovation, popularization and application of fishery technology and the management level of fishery organizations. Therefore, the quality of the labor force should not be ignored. In this study, both the quantity and the quality of the labor force are taken into account, which makes the calculation of the contribution of fishery technological progress closer to reality.

According to the formula of Hall and Jones [33], labor quality and labor quantity are combined to form a new labor input factor. E_it represents the average number of years of education for fisheries practitioners in the i area of the t period:

$$\mathrm{L}{^}_{\mathrm{it}}={\mathrm{G}}_{\mathrm{it}}{\mathrm{L}}_{\mathrm{it}}={\mathrm{e}}^{\mathsf{\omega }\left({\mathrm{E}}_{\mathrm{it}}\right)}{\mathrm{L}}_{\mathrm{it}}, {\mathrm{G}}_{\mathrm{it}}={\mathrm{e}}^{\mathsf{\omega }\left({\mathrm{E}}_{\mathrm{it}}\right)}$$

(8)

Among them, $\mathsf{\omega }\left({\mathrm{E}}_{\mathrm{it}}\right)$ is the educational return rate in Mincer’s income equation [34], which refers to the percentage of income increase for each additional year or stage of education, when ${\mathrm{E}}_{\mathrm{it}}$ = 0, there are $\mathsf{\omega }$(0) = 0, then $\mathrm{L}{^}_{\mathrm{it}}={\mathrm{e}}^{\mathsf{\omega }\left(0\right)}{\mathrm{L}}_{\mathrm{it}}$ = ${\mathrm{L}}_{\mathrm{it}}$ At present, there is still a lack of a recognized method of educational return rate in stages all over the world, so this paper quotes the estimated education data in China obtained from Professor Saccaro Psacharopoulos, an educational economist. According to the data, the educational return rate of China is 0.180, 0.134 and 0.151 in the primary, secondary and higher education stages [20], respectively. That is, the coefficient is 0.180, 0.134 and 0.151 when the years of education are 0–6 years, 6–12 years or greater, respectively. Assuming that the per capita years of higher education in a certain area is sixteen years, the per capita stock of human resources in that area is:

G_it = exp(0.18 × 6 + 0.134 × 6 + 0.151 × 4) = exp(2.488)

(9)

At present, China classifies its population’s education into six levels (i.e., completely illiterate and semi-illiterate, primary school, junior high school, senior high school, secondary vocational school and university [20]), and these six levels of education correspond to different years of education (i.e., 1 year, 6 years, 9 years, 12 years, 15 years and 16 years). Therefore, people who have had one year of education are considered as completely illiterate and semi-illiterate, people with 16 years of education are university graduates, and the others fall in the middle. Based on this, the average years of education of the rural labor force in China is calculated. L_it is expressed by the data of aquaculture personnel in China.

3.3. Variable Explanation and Data Source

Aquaculture output value (Y): The output value of aquaculture measured in 10,000 yuan. In order to reduce the impact of inflation on the results of this paper, this study is calculated at constant prices in 2012.

Aquaculture material input (K): Due to the lack of data, this paper calculates this parameter by first calculating the proportion of aquaculture intermediate materials consumed in fishery intermediate materials consumption, and then multiplying the proportion by fishery intermediate materials consumption. The intermediate consumption of aquaculture includes the consumption of raw materials, such as fry input, feed, purchased fishing gear and the expenses that need to be paid in the production process (i.e., repair fees and epidemic prevention fees). To avoid the influence of price change, the fishery material cost data were calculated at the constant price in 2012.

Labor input factor (L): The combination of aquaculture population’s own labor quality and the number of fishery employees is a new input factor of fishery labor.

Land input elements (n): The land input elements include freshwater aquaculture area and seawater aquaculture area.

Climate variable (Disa): The ratio of the disaster area to the fishery culture area in that year is regarded as a climate variable.

In this study, 261 sets of panel data from 29 provinces in China between 2012 and 2020 were used, and the data came from the China Fishery Statistical Yearbook, the China Rural Statistical Yearbook and the China Statistical Yearbook.

4. Results and Analysis

4.1. Regression Results

Using the input of aquaculture factors in China from 2012 to 2020, output value and other related statistical data and using Frontier4.1 software, the above Formula (1) is calculated through regression (as shown in

Table 2. Estimation results of production function parameters.

Variable	Coefficient	Standard Error	T Value
c	5.1013	2.9631	1.7216
t	−0.0231	0.0698	−0.33085
t square	0.0022	0.00515	0.42592
lnk	−0.0146	0.3234	−0.04521
lnL	0.494	0.7366	0.67062
lnN	−0.1466	0.5195	−0.28223
tlnK	−0.0135	0.01226	−1.10036
tlnL	−0.0106	0.01018	−1.03936
tlnN	0.02809	0.00768	3.6574
1/2lnklnk	0.1072	0.02805	3.8215
1/2lnLlnL	−0.0069	0.14463	−0.04752
1/2lnNlnN	−0.012	0.1324	−0.090823
1/2lnklnL	−0.0589	0.04368	−1.3487
1/2lnklnN	−0.0086	0.07271	−0.11846
1/2lnLlnN	0.04198	0.24199	0.17349
disa	−0.1247	0.20358	0.6127
γ	0.9991	0.03245	3.0788
LR	102.125
Observation measurement	261

). According to Table 2, there are three variables whose statistical values of T pass the test at the significance level of 1%, and five variables that pass the test at the significance level of 10%. The one-sided likelihood ratio LR = 102.125 rejects the original hypothesis at the significance level of 0.01, and the γ value is 0.9991, which indicates that there are technical inefficiency items in this model and that the stochastic frontier model can be used.

It can be seen from Table 2 that the coefficient of t² in the model is positive (indicating the existence of hard technological progress) and the regression coefficient of disa is negative (indicating that climate factors reduce the output value of aquaculture by 12.47%). The coefficient of lnLlnN is positive, indicating that the interaction between land and labor factors is positive, which also indicates that both labor and the aquaculture area have complementary advantages in aquaculture.

4.2. Calculation of Contribution Rate of Technological Progress

4.2.1. Calculation of Contribution Rate of Hard Technological Progress

According to Formula (2), the contribution rate of hard scientific and technological progress of aquaculture industry is calculated, and the analysis of the results is shown in

Table 3. Progress rate and contribution rate of aquaculture hard technology in China from 2012 to 2020.

Age	Hard Technological Progress Rate%	Neutral Technological Progress	Partial Technological Progress	Output Growth Rate%	Contribution Rate of Hard Technological Progress%
2012	6.552	−0.0209	0.0864	12.225	53.597
2013	6.597	−0.0187	0.0847	8.206	80.39
2014	6.644	−0.0165	0.0829	7.992	83.131
2015	6.794	−0.0143	0.0823	3.773	180.066
2016	6.898	−0.0121	0.0811	3.934	175.342
2017	7.112	−0.0099	0.081	8.432	84.345
2018	7.395	−0.0077	0.0817	5.401	136.908
2019	6.792	−0.0055	0.0734	−11.008	−61.699
2020	7.872	−0.0033	0.082	−9.089	−86.603
Average Value	6.962	−0.0121	0.0817	3.319	71.72

. From 2012 to 2020, the progress rate of aquaculture hard technology in China increased year by year. Hard technological progress can be divided into neutral technological progress and biased technological progress. Table 3 indicates that the neutral technological progress is negative while the biased technological progress is positive, which means that the current aquaculture hard technological progress in China is dominated by biased technological progress. This reflects that the input factors of aquaculture have value bias in the process of production or flow processing, and the progress of some technical factors will further increase the marginal output of production factors. Technological progress plays a technical role through its combination with material elements, and through technological progress, the marginal productivity of factor inputs is improved and the economic growth of aquaculture is promoted.

Table 3 also indicates that, from 2012 to 2020, the average contribution rate of hard technological progress to China’s aquaculture industry reached 71.720%, showing an upward trend from 2012 to 2018 with a large increase from 24.666% in 2012 to the highest point of 180.066% in 2015, and then falling back to 84.345% in 2017 until 2018. The contribution rate of aquaculture technological progress is spiraling because science and technology do not immediately show their promoting effect on aquaculture development, but rather are cyclical, long-term and lagging. In 2019 and 2020, the contribution rate of hard technological progress showed a negative value, mainly because of the COVID-19 epidemic and that Sino–US trade friction restricted aquaculture production and business activities. The production, transportation, processing and sales of fisheries were all affected, leading to negative fishery economic output value and the growth rate of fishery economic output value being negative.

4.2.2. Calculation of the Contribution Rate of Soft Technological Progress

According to Kumbhakar’s decomposition formula, the generalized scientific and technological progress rate can be divided into three items: hard technological progress rate, scale profit change and technological efficiency change rate. The latter two items represent the soft technological progress rate, and the calculation results are shown in

Table 4. General technological progress rate and contribution rate of fisheries in China from 2012 to 2020.

Age	Scale Reward Index	Rate of Change in Return on Scale	Technical Efficiency	Contribution Rate of Soft Technological Progress%	Generalized Technological Progress Rate%	Contribution Rate of Generalized Technological Progress%
2012	0.97	0.00721	0.9626	5.899	7.274	59.496
2013	0.979	0.00725	0.9634	9.805	7.401	90.195
2014	0.985	0.00728	0.9642	10.105	7.452	93.236
2015	0.991	0.0073	0.9649	21.206	7.594	201.272
2016	0.998	0.00732	0.9657	20.64	7.71	195.982
2017	1.001	0.00733	0.9664	9.526	7.915	93.871
2018	1.008	0.00734	0.9671	14.876	8.198	151.784
2019	1.036	0.00733	0.9678	−7.293	7.595	−68.992
2020	1.016	0.00731	0.9685	−8.813	8.673	−95.416
averge value	0.998	0.0073	0.96562	8.439	7.757	80.159

.

It can be seen from Table 4 that the scale reward index of China’s aquaculture industry from 2012 to 2020 (that is, the sum of elasticity of land, labor and materials) is close to and greater than one in recent years, indicating that the scale benefit level of China’s aquaculture industry has been increasing and that intensive aquaculture will bring scale benefits. In recent years, with the acceleration of urbanization in China, the loss of a rural labor force is serious, and aquaculture is facing the problem of an insufficient supply of an effective labor force. In addition, China’s land resources are scarce. Under this background, aquaculture not only needs to consider factor inputs but also needs to improve the utilization rate of input factors through technological progress. From 2012 to 2020, the average contribution rate of hard technological progress is 71.720%, that of soft technological progress is 8.439% and that of generalized technological progress is 80.159%. Therefore, the contribution of hard technological progress to the economic growth of aquaculture in China is dominant, while the contribution of soft technological progress is obviously smaller.

There is an urgent need to increase the contribution of soft technological progress in China aquaculture industry. As said by Denison [35], the importance of management decision-making and knowledge progress to economic growth should basically be the same, and we cannot just focus on the latter and ignore the former. The effective exertion and application of hard technology cannot be separated from the realization of the function of soft technology. Aquaculture subjects not only pay attention to the application of hard technology in aquaculture but also need effective coordination with soft technologies such as management, aquatic science and technology extension services, information technology and decision support so as to better improve the input–output efficiency of hard technology, promote the coordination among various input elements and achieve the optimal allocation of elements. Only when soft technology and hard technology work together can we effectively promote the development of aquaculture with high quality and efficiency.

Figure 1 is the trend chart of the general technological progress rate, the hard technological progress rate and the soft technological progress rate of China’s aquaculture industry from 2012 to 2020. Clearly, the general trend of China’s hard technological progress rate and its general technological progress rate is on the rise from 2012 to 2020, and the contribution of hard technological progress to aquaculture output and economic growth is dominant and increasing year by year, while the soft technological progress rate is staying at a low level. This research conclusion is consistent with Jiang Qijun’s viewpoint [36]; that is, the synergy degree of the fishery complex system in China is generally low, and improving the synergy degree of the fishery complex system is beneficial to promoting fishery economic growth. This is mainly related to the fact that China’s aquaculture business entities are mainly scattered small and medium-sized fishermen, with a small aquaculture scale and extensive management, neglecting organizational innovation, brand operation and marketing construction, and lacking large-scale aquaculture business entities (such as large professional households and leading enterprises). On the other hand, the reason for this could be the decline in the number of fishery scientific research extension institutions, the number of fishermen’s training periods and the number of trainees (person-times) (as shown in

Table 5. Promotion of fishery science and technology in China from 2012 to 2020.

Particular Year	Number of Fishery Scientific Research Institutions	Number of Training Periods for Fishermen	Number of Trainees (Person-Time)	Number of Published Scientific Papers	Patent of Invention
2012	14,711	40,183	2,953,426	2335	184
2013	14,728	39,998	2,867,116	2682	254
2014	14,755	34,987	2,254,971	2860	300
2015	14,398	32,153	2,086,460	2875	275
2016	13,463	18,057	1,367,249	2787	309
2017	12,305	18,057	1,367,249	2852	221
2018	11,976	16,702	1,018,636	2739	333
2019	11,705	13,840	992,334	2752	315
2020	11,373	13,775	910,037	2857	374

The data comes from China Fishery Statistics Yearbook [2].

and Figure 2 and Figure 3). The number of published scientific papers and invention patents are the representatives of hard technology [37]. The number of published scientific papers and invention patents in China is on the rise, which is consistent with the conclusion that the progress rate of hard technology is increasing year by year. The main function of fishery promotion agencies is to carry out technology promotion; introduce, demonstrate and popularize new technologies; and provide technical training for fishery practitioners (thus representing soft technologies). The number of fishermen’s training sessions and the number of trainees has shown a downward trend in recent years, which is consistent with the low progress rate of soft technologies.

5. Conclusions and Recommendations

Through the above analysis, we can find that the contribution rate of technological progress in China’s aquaculture industry from 2012 to 2020 is 80.159%, of which the contribution rate of hard technological progress is 71.720% and the contribution rate of soft technological progress is only 8.439%. It can be found that the progress of hard technology plays a leading role in the economic growth of China’s aquaculture, while the role of soft technology is minimal. However, the progress of soft technology is also an important factor that cannot be ignored for a sustainable aquaculture sector. In recent years, China’s aquaculture industry has attached great importance to the innovation of hard technologies (such as research and development of new technologies) but ignored the role of soft technologies in aquaculture economic growth. The modern fishery competition has changed from the competition among products to the competition in the whole industrial chain, developing towards integration, recycling and integration, and transforming into comprehensive formats, such as fishery complexes, ecological recycling bodies and industrial parks. This requires the integration and innovation of hard technological progress and soft technological progress, as well as the coordination of science, technology, culture and ecology.

We must not only pay attention to the contribution of hard technological progress to the development of aquaculture, but also pay attention to the contribution of soft technological progress to the development of aquaculture industry: first, fishery research and promotion institutions should strengthen cooperation with local leading enterprises, regularly carry out fishery science popularization and technical service activities, strengthen the cultivation of modern new fishery management entities and transform and upgrade from pure aquaculture production to “fishery industry and trade” integration. Second, aquatic technology promotion agencies should provide fishermen with seedlings, feed, disease prevention and control technology, new breeding technology, training of new concepts, etc., to promote the effective synergy between hard technological progress and soft technological progress on aquaculture economic growth.