Agricultural Green Total Factor Productivity in Shandong Province of China

Peng, Yuanxin; Chen, Zhuo; Lee, Jay Article Agricultural Green Total Factor Productivity in Shandong Province of China German Journal of Agricultural Economics (GJAE) Provided in Cooperation with: Gesellschaft für Wirtschaftsund Sozialwissenschaften des Landbaues e.V. (GEWISOLA) Suggested Citation: Peng, Yuanxin; Chen, Zhuo; Lee, Jay (2024) : Agricultural Green Total Factor Productivity in Shandong Province of China, German Journal of Agricultural Economics (GJAE), ISSN 2191-4028, TIB Open Publishing, Hannover, Vol. 73, Iss. 2, pp. 1-25, https://doi.org/10.52825/gjae.v73i2.1351 This Version is available at: https://hdl.handle.net/10419/305175 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ German Journal of Agricultural Economics Vol. 73 (2024), No. 2, 1-25 Original Research Article https://doi.org/10.52825/gjae.v73i2.1351 © Authors. This work is licensed under a Creative Commons Attribution 4.0 International License Submitted: 22 Aug. 2022 | Accepted: 04 Apr. 2024 | Published: 31 May 2024 Agricultural Green Total Factor Productivity in Shandong Province of China Yuanxin Peng 1, Zhuo Chen 2, and Jay Lee 3* 1 Zaozhuang University, Zaozhuang, Shandong, China 2 Case Western Reserve University, Cleveland, and Harrington Heart and Vascular Institute, University Hospitals, Cleveland, OH, USA 3 Kent State University, Kent, USA *Correspondence: Jay Lee, [email protected] Abstract: Sustainable development of agriculture has an important impact on both society and economy. In order to understand the patterns of spatio-temporal variation and the factors influencing agricultural green total factor productivity (AGTFP), this paper used Shandong province of China as a case study. Utilizing the SBM-DEA and Malmquist models, along with panel regression methods, the study analyzes AGTFP based on data from the Shandong Statistical Yearbook (2009-2019). The results showed that: (1) the AGTFP in Shandong province was smaller than the total factor productivity when not considering the undesirable output, and the AGTFP in most regions of Shandong province needed to be improved. (2) The AGTFP of Shandong province showed an annual rising trend, especially in the eastern and northern regions. (3) In addition to the levels of technology and management, the industrialization and level of personal development of farmers is also shown to have impacted on AGTFP. Recommendations include adopting advanced technologies, enhancing land management, promoting tertiary sector development, expanding agricultural processing, and improving farmer skills through education and training to boost AGTFP to achieve a sustainable agricultural economy. Keywords: AGTFP, SBM-DEA Model, Malmquist Model, Center of Gravity Model, Coefficient of Variation, Panel Regression 1 Introduction Since the reform and opening up in 1978, China's rural economy had developed rapidly, and the output of various agricultural products had increased significantly. In recent years, the output of major agricultural products such as grain, oil, vegetables, fruits, meat, poultry and eggs were among the highest in the world (Rmlt, 2019; Chinairn, 2020). Shandong province is located in the east coast of China, with its excellent geographical location (see Appendix A), suitable climatic conditions and a developed agricultural economy. It was often ranked as the first in China in terms of gross output value of agriculture, added value of agriculture, export value among other indicators (Song et al., 2012). Shandong was also ranked as the third in grain crop yield and sown area of vegetables, the first in total fruit production, and the first in total output value of animal husbandry (China Statistical Yearbook, 2009-2019). Agricultural development requires the use of a large amount of chemical fertilizers, and such use in Shandong province has long been the second highest in China, second only to Henan province (China Statistical Yearbook, 2009-2019). Grain, vegetable, and other crops produce massive amount of straw and livestock breeding produces a great deal waste. 1 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 Excessive use of chemical fertilizer, waste from livestock breeding and inappropriate disposal of straw cause serious non-point source pollution in rural area. Agricultural non-point source pollution is the pollution generated in agricultural production activities that pollutants enter water through farmland surface runoff, soil flow, farmland drainage and underground leakage (Ma et al., 2009). It is estimated that agricultural non-point source pollution accounts for one-third of the total water pollution in China (Li, 2022). Among them, CODcr, TN, and TP accounted for 44%, 57% and 67%, respectively, of the total discharge of each pollutant (Huang et al., 2012). CODcr is the chemical oxygen consumption measured by using potassium dichromate (K2Cr2O7) as oxidant, namely the dichromate index. TN (Total Nitrogen) is the total amount of nitrogen present in soil or water. It is calculated as the milligrams of nitrogen per liter of water. TN is commonly used to indicate the degree of nutrient pollution in water bodies. The higher the TN value, the more severe the water quality pollution. TP (Total Phosphorus) is the total content of phosphorus in soil or water. It is one of the indicators used to measure the level of water pollution. A higher TP value indicates a higher degree of water quality pollution. Therefore, it is necessary to study issues related to agricultural sustainable development, especially the green agricultural productivity. This is because improving green agricultural efficiency can reduce inputs of agricultural production resources and the generation of pollutants, thereby promoting agricultural sustainability. Considering that the agricultural development of Shandong province plays a very important role in China, it has great significance to study the patterns of spatio-temporal variation and mechanism associated with AGTFP in Shandong province. Studying AGTFP in the temporal dimension allows us to understand its patterns of change over time, while studying it in the spatial dimension enables us to understand its spatial distribution characteristics. We first calculated agricultural non-point source pollutants and used them as the undesirable output. We selected variables such as agricultural GDP, agricultural labor force, the total power of agricultural machinery, arable land area, and irrigated land area for the calculation of AGTFP in Shandong province. Then, the spatiotemporal variation patterns of AGTFP in Shandong province were analyzed based on the calculated results. Finally, panel data analysis was conducted to explore the mechanisms of changes in AGTFP. The rest of the paper is organized as follows: Section 2 is primarily a literature review that aims to enhance readers' understanding of diverse perspectives on agricultural total factor productivity (agricultural TFP) and AGTFP. Section 3 outlines the selection of indicators and data characteristics. Section 4 describes the calculation methods. Section 5 analyzes the calculation results. Section 6 discusses the implications and significance of the findings. Finally Section 7 encompasses the conclusion and recommendations. 2 Literature Review Agricultural growth decomposes growth into total input use and total factor productivity (TFP). In particular, TFP has become the primary source of agricultural growth worldwide (USDA, 2012). To a large extent, agricultural modernization is the process in which TFP's contribution to agricultural economic growth is expected to continue rising. Since agricultural TFP is of great significance to agricultural development, many scholars have studied the trends and factors of agricultural TFP from different perspectives: (1) Studies at Different Spatial Scales Some scholars have examined agricultural TFP at various spatial scales, including global, continental, and national levels. For instance, Fuglie (2015) conducted an analysis of global agricultural TFP for the years 1961-2012. The results suggested that the rate of agricultural TFP 2 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 growth had accelerated in recent decades. Alhassan (2021) conducted a study using data from 38 countries in sub-Saharan Africa (SSA) to investigate the impact of agricultural TFP on environmental degradation. The findings revealed a U-shaped relationship between agricultural TFP and carbon dioxide emissions in SSA. (2) Studies from Perspectives of Different Influencing Factors Many scholars have studied the effects of different influencing factors on agricultural TFP. For instance: Li et al. (2021) investigated the relationship between China's rapid urbanization and agricultural TFP. The results revealed a U-shaped relationship between urbanization and agricultural TFP. Through a study of agricultural TFP in 15 countries in South Asia and Southeast Asia, Liu et al. (2020) discovered that human capital had a positive influence on the growth of agricultural TFP. Espoir et al. (2021) in their study of agricultural TFP in Africa. They highlighted that good governance can play a pivotal role in enhancing agricultural productivity. Yang et al. (2019) found rural human capital positively contributes to the local agricultural TFP, while adjustments in crop structure significantly restrain the increase in local agricultural TFP levels. In the process of agricultural development, the extensive use of chemical fertilizers and pesticides can lead to environmental pollution. Additionally, the large amounts of manure generated by livestock and poultry breeding also contribute to environmental degradation. The discrepancy between TFP calculations that do not consider the losses caused by environmental pollution and the actual TFP can easily lead decision-makers to develop policies that are unfavorable to green development. To promote the harmonious development of agriculture and the environment, scholars have incorporated environmental pollution factors into the analysis of agricultural TFP, resulting in the concept of AGTFP (Xu et al., 2020). A few scholars have studied AGTFP. Yang et al. (2022) discovered a significantly positive relationship between rural human capital and AGTFP in their study of AGTFP across 28 provinces (cities and autonomous regions) in China. Han et al. (2018) identified that planting structure has a slight negative effect on AGTFP in their analysis. Using panel data from 30 Chinese provinces, Wang and Xie (2022) conducted an analysis of the relationship between human capital and AGTFP. The results indicated that the significant improvement in the quality of human capital notably influences the growth of AGTFP in China. Wang and Xiao (2022) found that the massive population migration from rural to urban areas during the urbanization process results in a continuous deterioration of agricultural green productivity. Liang and Long (2015) conducted an analysis of the factors influencing the growth of AGTFP in 31 provincial-level administrative regions of China. They found that the impact of agricultural fiscal expenditures on AGTFP was not particularly significant, and the advancement of industrialization hindered the increase in AGTFP growth rate. Yang et al. (2019) examined the spatial variation of AGTFP and its driving factors and found that the impact of economic development level, agricultural structure, and financial support for agriculture on AGTFP exhibited regional variability. Some scholars have also studied AGTFP in Shandong province. For example, Zhang and Liu (2015) used the C2R model in DEA to evaluate the agricultural productivity in Shandong province. C2R model is a model built on the premise of constant return to scale, which is applicable to the situation where the input increase in a certain proportion and the output also increase in proportion to the input. However, agricultural production did not fit this situation. Jiao (2013) analyzed the agricultural productivity of Shandong province. Jiao (2013) mainly adopted industrial pollutant discharges to represent the undesirable output of agriculture. However, this method was deficient as industrial pollutants often had no connection with agricultural production. In order to better understand the pattern and determinants of AGTFP in Shandong province, we employed the Slacks-Based Measure (SBM)-DEA method and other approaches to calculate and analyze AGTFP. This paper considered agricultural non-point source pollutants as the 3 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 undesirable output and conducted statistical analysis of AGTFP in Shandong province. We treated the undesirable output as an input to make a dynamic comparison of AGTFP. 3 Index Selection and Data Description 3.1 Index Selection of AGTFP DEA is a research method for multi-factor input and output evaluation. When performing calculations, it necessitates the selection of input and output variables. This approach combines input and output variable data from readily available statistical sources. Output variables were agricultural Gross Domestic Product (GDP) and agricultural non-point source pollutants in each city. Input variables were agricultural labor population, the total power of agricultural machinery, cultivated land area, irrigated land area and chemical fertilizer consumption. (1) Agricultural GDP: The value of agricultural economic output is expressed by the added value of agriculture, forestry, animal husbandry and fishery in units of 10,000 Yuan (CNY). To eliminate the impact of inflation, the output values were converted into constant price in 2008 based on the GDP deflator in different years and different local cities and municipalities. (2) Pollutant: The calculated agricultural non-point source pollutants were used, with the unit being ton. (3) Agricultural labor population: Agricultural labor refers to the number of individuals engaged in agricultural industry, the unit was ten thousand. (4) The total power of agricultural machinery: the unit is kilowatt. Agricultural machinery refers to equipment such as tractors, harvesters, and planters that are used in agricultural production. Hong et al. (2022) utilized this indicator in their study on the impact of digital inclusive finance and optimization of agricultural industry structure on AGTFP. (5) Cultivated land area: Agricultural production required the occupation of land, we chose cultivated land area as the input, with the unit being hectare. (6) Irrigated land area: Due to the lack of irrigation water data, the actual area of irrigation land was used instead, and the unit was 1,000 hectares. (7) Chemical fertilizer consumption: A mass of chemical fertilizer was used in agricultural production. We used the fertilizer after converted to pure volume, and the unit was ton. 3.2 Factor Selection in Panel Regression We selected influencing factors in a panel regression model by considering the interplay among various factors affecting AGTFP, while also taking into account data availability and factors employed in previous research studies. The independent factors in the model include urbanization (Fang et al., 2021), agricultural industrial structure (Liu et al., 2021; Liu, 2018), industrialization (Fang et al., 2021), the influence of government on agriculture (Liu et al., 2021; Yang et al., 2022), farmers’ characteristics (Ye et al., 2023), economic development level (Wang and Wang, 2017) and distance from port (Li et al., 2022). The explanations of each factor are as follows: (1) Urbanization ratio: In the process of urbanization, part of the rural population was transferred to cities because of education, work, medical care and other reasons, which led to serious aging in rural areas and the abandonment of land. This would directly affect the output of agriculture. (2) Agricultural industrial structure: It was represented by the proportion of grain crop area to total crop sown area, which indicated adjustment of agricultural structure. Because the benefits produced by food crops and cash crops are different, it would have an impact on AGTFP. 4 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 (3) Industrialization: The proportion of value-added by the secondary sector to the GDP in each city was utilized as a measure of the level of industrialization. Industrialization has the potential to attract labor migration from agriculture, with many young and middle-aged individuals entering factory employment, resulting in a shortage of labor in rural areas. This phenomenon can have a negative impact on agricultural TFP. However, the industrial sector can also contribute to the improvement of agricultural TFP by manufacturing advanced agricultural machinery for use in agricultural production. (4) The influence of government on agriculture: We used proportion of fiscal expenditure on agriculture, forestry and water resources to the fiscal expenditure. This factor reflected the state of government support for agriculture. The more the government invested in agriculture, the more agricultural scientific research results, and the higher the agricultural technical efficiency there would be. (5) Personal development of farmers: There was no directly related data for this indicator. We used the proportion of farmers' expenditure on education, culture and entertainment in their annual consumer expenditure. (6) Economic development level: Per capita GDP was used instead, and the unit was ten thousand CNY. (7) Distance from port: Qingdao port has been one of the famous ports in the world. This paper intended to measure the influence of the port on AGTFP by using the distance between each city and Qingdao. The distance from each city to Qingdao was calculated based on longitude and latitude coordinates. The distance of Qingdao to itself was calculated by using the area of Qingdao, calculating the average radius and it represented the distance of Qingdao. 3.3 Data Description The data covers 17 cities in Shandong province, with a time span from 2008 to 2018, and each observation variable consists of 187 values. It is important to note that in China, the next level of administrative units below the provincial level is the city, which includes both urban and rural areas. The descriptive characteristics of these data can be found in Appendix B. Due to the distribution of these data across 17 cities and spanning 11 years, there are significant differences between the variables. The main variables showing an increasing trend include agricultural GDP, per capita GDP, irrigated land area, urbanization ratio, agricultural industrial structure, the influence of government on agriculture, and personal development of farmers. The variables showing a decreasing trend include agricultural labor population, total power of agricultural machinery, cultivated land area, fertilizer usage, pollutant quantity in agriculture, and industrial structure of the secondary sector. The variable that remains unchanged is the distance of each city from the Qingdao port. 4 Methods 4.1 Calculation of Agricultural Pollutant Discharge This paper used the unit survey evaluation method to calculate the discharges of agricultural pollutants. The pollution unit was the non-point source pollution unit, which was the smallest independent unit that produced non-point source pollution. This could be measured statistically, such as fertilizer, crop straw, livestock and poultry breeding. The coefficients involved in the calculation of pollutant discharge were mainly adopted from Liang (2009) and Lai (2003). For detailed information, please refer to the papers by Liang and Lai. Calculation method of pollutant discharge was as the following: 5 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 𝐸=∑𝐸𝑈𝑖𝜌𝑖𝑘(1−𝜂𝑖) 𝑖 (1) where E is the discharge of agricultural non-point source pollutants, i.e., CODCr, TN, and TP, i is pollution unit, 𝐸𝑈𝑖 is the number of agricultural pollution unit i , 𝜌𝑖𝑘 is the pollution intensity coefficient of pollutant k in agricultural pollution unit i , which is the amount of pollutants produced by a pollution unit. The pollutant indexes considered in this paper were the production of TN, TP and CODCr (Lai et al., 2004). i  is coefficient of utilization efficiency of agricultural pollution unit i . 1-𝜂𝑖 is coefficient of run-off of agricultural pollution unit i. Nitrogen fertilizer and phosphorus fertilizer in chemical fertilizers are important sources of TN and TP in agricultural non-point source pollution. The calculation methods for these pollutants are as follows: The consumption of fertilizer is calculated by the usage of nitrogen and phosphate fertilizers after conversion to pure volume. The fertilizer after conversion to pure volume refers to the amount of nutrients of each fertilizer summed by the mass percentage of N, P2O5 and K2O. Thus, nitrogen fertilizer after conversion becomes the amount of TN; phosphorus fertilizer after conversion is the amount of P2O5. Therefore, the content of P in P2O5 is about 43.66%. The amount of TP is the product of the phosphate fertilizer after conversion to pure volume and 43.66%. The loss of TN, TP can then be derived by multiplying the amount of TN, TP and their loss rates respectively. The nitrogen loss rate is 20% and the phosphorus loss rate is 7% in Shandong province. Livestock and poultry farming is another significant source of agricultural non-point source pollution. The calculation methods for the pollutants generated from livestock and poultry farming are as follows: The production of pollutants from livestock and poultry breeding = the amount of livestock and poultry kept at the end of the year × the excretion coefficient of pollutants from livestock (Table 1) and poultry breeding sources × the excretion loss rate. The excreta loss rate of livestock and poultry in Shandong province was 27.6% CODCr, 24.4% TN and 21.2% TP. Table 1. Annual excretion coefficient of pollutants from livestock and poultry (kg/unit) Pollution unit i CODCr TN TP cattle 401.500 61.100 10.070 swine 47.880 4.510 1.700 sheep 4.400 2.280 0.450 poultry 1.165 0.275 0.115 Source: coefficients adopted from Liang (2009) and Lai (2003) Solid waste generated from agricultural production is also an important source of agricultural non-point source pollution. The calculation methods for this type of pollutant are as follows: Farmland solid waste is mainly crop straw. The calculation of farmland solid waste involves the consideration of factors such as the crop straw to grain ratio, pollution production coefficient, and emission coefficient. Since there are various types of vegetables with different waste proportions, this study assumed an average solid waste proportion of 0.51 for vegetables. See Table 2, Table 3, Table 4, and Table 5 for the specific coefficients for calculation. In the context of Shandong, taking into account the proportion of straw utilization and nutrient loss, it was found that the loss proportions for CODcr, TN, and TP were 11.57%, 10.39%, and 8.61% respectively. 6 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 Table 2. Main crop straw grain ratio type paddy wheat corn bean potato Oil crops straw: grain 0.970 1.030 1.370 1.710 0.610 2.260 Source: coefficients adopted from Liang (2009) and Lai (2003) Table 3. Pollution production coefficient of different crop straw unit pollution production coefficient (10-3 t/t) CODcr TN TP paddy 5.630 5.820 0.420 wheat 6.390 5.150 0.900 corn 11.230 10.690 2.390 bean 17.610 22.230 2.240 potato 2.260 1.830 0.670 Oil crops vegetable 20.570 5.100 45.430 0.920 3.060 0.450 Source: coefficients adopted from Liang (2009) and Lai (2003) Table 4. Straw utilization ratio in Shandong (%) fertilizer fodder fuel raw material incineration stack total 23.600 31.000 19.600 6.300 5.800 13.700 100.000 Source: coefficients adopted from Liang (2009) Table 5. Straw utilization and nutrient loss ratio (%) nutrient fertilizer fodder fuel raw material incineration stack CODcr 20.000 0.000 0.000 0.000 0.000 50.000 N 15.000 0.000 0.000 0.000 0.000 50.000 P2O5 5.000 0.000 0.000 0.000 10.000 50.000 Source: coefficients adopted from Liang (2009) and Lai (2003) 4.2 Dimension Reduction of Pollutants Due to the presence of three pollutants - CODcr, TN and TP, and that DEA analysis requires the decision-making unit to be more than twice the sum of input variables and output variables, the dimension of pollutants needed to be reduced. Here principal component analysis (PCA) method was used for dimensionality reduction. Subsequently, the coefficient from the Component Score Coefficient Matrix was used as the weight to calculate the sum of CODcr, TN and TP, and the sum was taken as the undesired output. After the three pollutants were processed by using PCA, the contribution of variance in data by the first principal component was above 95%, so the first principal component can be used to replace the three pollutants. With this, the formula was restructured to be: 𝐹𝐴𝐶=𝑎1∗𝑋1+𝑎2∗𝑋2+𝑎3∗𝑋3 (2) where FAC was the pollutant after dimensionality reduction, a1, a2, a3 were component score coefficients and X1, X2 and X3 are the three pollutants: CODcr, TN and TP. 7 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 4.3 SBM (Slacks-Based Measure)-DEA Approach The SBM model is a type of DEA model. Compared to other DEA models, the SBM model allows for the measurement of efficiency changes under non-expected output constraints (Tan and Liu, 2022). Therefore, it can better reflect the essence of efficiency evaluation than other models (Tu and Liu, 2011). Tone (2001) proposed and developed an SBM-DEA model. In the SBM-DEA model: Suppose production systems have n decision making units, DEA analysis would be an economic system or a process (one unit), which would be considered as an entity. Within a certain possible extent, it works by putting a number of factors of production and output of a certain number of "products”. Such entities (units) are called decision-making units (DMUs). Each unit contains three vectors of input: desirable output and undesirable output. They are denoted as m xR , 1gs yR , 2bs yR . Define the matrix of X , g Y , b Y where [𝑋]=[𝑥1,⋯,𝑥𝑛]𝑇∈𝑅𝑚×𝑛, [𝑌𝑔]=[𝑦1𝑔,⋯,𝑦𝑛𝑔]𝑇∈𝑅𝑠1×𝑛, [𝑌𝑏]=[𝑦1𝑏,⋯,𝑦𝑛𝑏]𝑇∈𝑅𝑠2×𝑛, X >0, g Y >0, b Y >0. Define the production possibility set 𝑝 as: 𝑝={(𝑥,𝑦𝑔,𝑦𝑏)|𝑥≥𝜆𝑥,𝑦𝑔≤𝜆𝑌𝑔,𝑦𝑏≤𝜆𝑌𝑏,𝜆≥0} (3) Then the SBM model based on variable return scale is expressed by formula (2): 𝑝∗=𝑚𝑖𝑛 1−1 𝑚∑𝑠𝑖− 𝑥𝑖𝑜 𝑚 𝑖=1 1+ 1 𝑠1+𝑠2[∑𝑠𝑟𝑔 𝑦𝑟0 𝑔+∑𝑠𝑟𝑏 𝑦𝑟0 𝑏 𝑠2 𝑖=1 𝑠1 𝑖=1 ] (4) where 𝑠 is the slacks of input and output, 𝜆 is weight vector, objective function 𝑝∗ with respect to 𝑠−, 𝑠𝑔, 𝑠𝑏is strictly decreasing and 0≤𝑝∗≤1. For a particular decision unit, if and only if 𝑝∗=1 and 𝑠−, 𝑠𝑔, 𝑠𝑏 are 0, the comprehensive efficiency is effective. When 𝑝∗<1 or 𝑠−, 𝑠𝑔, 𝑠𝑏are not complete zeroes, it indicates that the decision unit is inefficient, and the technical efficiency or scale efficiency is also invalid, so there is a need to improve the input and output. 4.4 Malmquist Method The SBM-DEA model is able to perform statistical analysis on cross-sectional data, but it does not measure the temporal trend of AGTFP, and hence cannot make a dynamic comparison. The Malmquist index can be used to solve this problem by combining cross-sectional data analysis with time series data analysis. It can decompose this productivity change into technical change and technical efficiency change (Sathye, 2002). 8 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 The Malmquist index can be decomposed into effch and techch. It can be seen from Table 8 that the changes in AGTFP were mainly caused by techch, indicating that agricultural technological progress led to the improvement of AGTFP. Agricultural technological progress includes measures such as improving agricultural machinery levels, using high-quality seeds, and implementing other technological advancements in agricultural practices. This result was consistent with the research conclusion in Sheng et al. (2020) that “It is widely believed that technological progress had played an essential role in contributing to the rapid productivity growth in China’s agricultural sector”. effch represents the combined efficiency of agricultural management level and input factors. Jinan, Zibo, Yantai and other 7 regions had been improved, but the remaining 10 regions were not efficient. This was in line with the current situation of low overall agricultural efficiency, extensive agricultural management and large numbers of farmers unwilling to engage in agricultural production. Table 8. Decomposition of AGTFP of cities city effch techch agtfp Jinan 1.000 1.055 1.055 Qingdao 0.991 1.050 1.041 Zibo 1.015 1.048 1.064 Zaozhuang 0.988 1.044 1.031 Dongying 0.990 1.061 1.050 Yantai 1.000 1.053 1.053 Weifang 0.998 1.059 1.056 Jining 1.000 1.043 1.043 Taian 0.995 1.043 1.038 Weihai 1.000 1.046 1.046 Rizhao 1.008 1.039 1.048 Laiwu 0.993 1.045 1.038 Linyi 0.999 1.041 1.039 Dezhou 0.995 1.050 1.045 Liaocheng 0.999 1.046 1.045 Binzhou 1.010 1.058 1.069 Heze 0.988 1.056 1.043 Source: data derived from the decomposition of the Malmquist index 5.3.2 Comparison Analysis of AGTFPT Trends Across Cities Table 9 shows the AGTFP data of all regions in Shandong province from 2009 to 2018. Table 6 presents AGTFP calculated using the SBM method, which allows for the analysis of AGTFP across different cities in the same year. On the other hand, Table 9 displays AGTFP calculated using the Malmquist Index method, providing a convenient means to analyze the temporal trends of AGTFP within the same region. In light of the temporal shifts, it is evident that the AGTFP in every city had been experiencing a consistent and progressive increase. In terms of the change characteristics of AGTFP, three distinct types can be discerned: (1) Jinan, Zibo, Weifan, Jining, Rizhao, Liaocheng, Binzhou, and Heze were keeping relatively high growth rates，AGTFP were greater than 1 annually. (2) In Zaozhuang, Laiwu and Linyi, AGTFP gradually developed from low efficiency to high efficiency, and changed from less than 1 to higher than 1. (3) The remaining 6 cities were experiencing fluctuating growth. Although the efficiency was less than 1 in some years, the overall efficiency was also constantly improving. 15 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 Table 9. The trend of AGTFP in 17 cities of Shandong province City 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Jinan 1.028 1.026 1.026 1.038 1.006 1.028 1.035 1.191 1.103 1.084 Qingdao 1.014 0.996 1.036 1.027 1.032 1.041 1.034 1.140 1.036 1.061 Zibo 1.017 1.033 1.028 1.024 1.028 1.029 1.044 1.319 1.083 1.064 Zaozhuang 0.972 0.996 0.997 1.017 1.016 1.016 1.024 1.174 1.068 1.044 Dongying 1.021 1.021 1.018 1.015 0.968 1.015 0.972 1.225 1.151 1.127 Yantai 1.021 1.024 0.997 1.038 1.025 1.035 1.041 1.211 1.023 1.135 Weifang 1.007 1.012 1.014 1.045 1.027 1.040 1.050 1.204 1.073 1.108 Jining 1.021 1.013 1.022 1.038 1.027 1.043 1.044 1.104 1.067 1.053 Taian 0.999 1.011 0.997 1.015 1.012 1.008 1.015 1.143 1.141 1.050 Weihai 1.030 1.019 1.006 1.029 0.985 1.033 1.075 1.032 1.033 1.236 Rizhao 1.019 1.039 1.013 1.026 1.048 1.043 1.165 1.056 1.061 1.016 Laiwu 0.992 0.989 1.008 1.045 1.017 1.019 1.012 1.251 1.033 1.034 Linyi 0.999 0.985 1.008 1.022 1.006 1.020 1.034 1.339 1.004 1.019 Dezhou 1.038 1.029 1.023 1.050 1.010 1.038 1.051 0.864 1.243 1.143 Liaocheng 1.040 1.035 1.038 1.047 1.010 1.042 1.051 1.062 1.049 1.077 Binzhou 1.019 1.017 1.032 1.033 1.023 1.007 1.038 1.348 1.037 1.176 Heze 1.020 1.030 1.028 1.028 1.030 1.025 1.037 1.158 1.032 1.050 Source: AGTFP calculated using the Malmquist Index method 5.4 Analysis of Center of Gravity Shift in AGTFP Within Shandong Province The latitude and longitude coordinates of each city in Shandong were obtained from the website (Jingweidu, 2020). iM was the AGTFP calculated by using Malmquist method. The CoG model was used to calculate the CoG of AGTFP in Shandong province from 2009 to 2018, which could reflect the spatial changes in the trajectory of the CoG of AGTFP in Shandong and could reveal their spatial pattern. The CoG is plotted in Figure 6. It can be seen from the figure that the center of gravity was 118.16°E and 36.38°N in 2009, and 118.20°E and 36.41°N in 2018. When compared with the geometric center of Shandong province (118.14°E,36.33°N) (Li, 2019), the centers were both on the east and north. This indicates that the AGTFP in eastern Shandong was higher than that in western cities, and the AGTFP in northern cities was higher than that in southern cities. In addition, the CoG generally had a tendency to shift eastward and northward, indicating that regional differences were increasing. 16 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 Figure 6. Trajectory of center of gravity for AGTFP in Shandong (horizontal axis represents longitude coordinates and vertical axis represents latitude coordinates) Source: data derived from calculations by the COG model 5.5 Evolution Characteristics of Regional Difference As can be seen from Figure 7, the CV of AGTFP in 2009 was 0.017. CV values had increased rapidly after 2014, and reached the maximum in 2016, with a value of 0.122. In 2017 and 2018, the CV values of AGTFP were reduced to 0.059 and 0.060. It can be seen from the figure that although Shandong’s AGTFP fluctuated, the overall trend was gradually increasing. This corresponded with the trajectory of Shandong AGTFP, indicating that the AGTFP growth rate difference in Shandong was increasing, and the increase of AGTFP in the eastern and northern regions was relatively large. Figure 7. Coefficient of Variation of AGTFP in Shandong province Source: data derived from calculations by the CV model 5.6 Analysis of the Factors Affecting AGTFP Besides agricultural technology and agricultural management level, AGTFP was also affected by urbanization level, personal development of farmers, agricultural structure and other factors. In order to identify the factors affecting Shandong’s AGTFP, this paper used panel data with a regression analysis of the factors that may affect AGTFP. 17 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 Figure 1, Figure 2 and Figure 3 indicate that AGTFP seemed to have spatial correlation. By employing the global Moran's I and local Moran's I analysis methods (Zhang et al., 2020a; Zhang et al., 2020b), we observed that only in 2018 there was a certain degree of spatial autocorrelation. No significant spatial autocorrelation coefficients in the rest of the years. Therefore, our regression analysis did not consider the spatial characteristics of that spatial distribution. The meanings of each variable can be found in Table 10. The magnitude of the regression coefficients in Table 11 indicates the extent of the independent variables’ impact on the dependent variable and does not imply causality. Correlation analysis and collinearity diagnostics were conducted among the independent variables, and no correlations exceeding 0.5 were identified. Furthermore, there was no evidence of multicollinearity among the independent variables. Hausman tests for both fixed effects and random effects were performed on the data, p=0.0003, the p value was less than 0.5 meaning fixed effects was preferred. It can be seen from Table 11 that the p values of urban, agstr, PGDP and fina were all greater than 0.1 which failed the significance test, indicating that these factors had no significant impact on AGTFP. dist was ignored directly and did not participate in calculation, indicating that distance to the port had no influence on Shandong AGTFP. The p value of ind was 0.000, indicating that the industrialization had a significant negative effect on AGTFP. This suggests a strong correlation between the two variables. The p value of pd was 0.078, indicating that personal development of farmers had a certain positive effect on AGTFP, indicating there is a certain level of correlation between the two variables. This is consistent with the conclusion in Zuo (2019) that “agricultural human capital and agricultural total factor productivity are significantly positively correlated”. Table 10. Description of the variables in regression analysis Variable Variable Description agtfp Agricultural Green Total Factor Productivity urban Urbanization level ind Industrialization agstr Agricultural operation structure PGDP Per capita GDP fina Proportion of fiscal expenditure on agriculture, forestry and water resources to the fiscal expenditure pd The proportion of farmers' expenditure on education, culture and entertainment in their annual consumption expenditure dist The distance from Qingdao cons Constant term of the regression equation Source: this table was compiled based on the variables used in the regression analysis of the study. Table 11. Regression results of the factors affecting AGTFP agtfp Coef. Std.Err. t P >t [95% conf. Interval] urban ind agstr PGDP fina pd dist cons -0.086 -1.170 0.048 0.092 -0.022 0.571 0. 000 1.590 0.090 0.246 0.114 0.224 0.338 0.322 (omitted) 0.220 0.950 4.760 0.420 0.410 0.060 1.770 7.220 0.341 0.000 0.674 0.681 0.949 0.078 0.000 -0.235 -1.578 -0.140 -0.279 -0.582 0.038 1.225 0.063 -0.763 0.236 0.463 0.539 1.105 1.954 F = 2.18 prob > F = 0.0079 Source: data obtained from the calculations using apanel regression model 18 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 6 Discussion This paper presents an analysis of AGTFP in Shandong Province, quantifying pollutants from fertilizers, livestock, and crop waste in agricultural non-point source pollution. Important factors were also determined by regression analyses. Our results indicate: (1) Based on the AGTFP data, approximately half of the cities had an AGTFP value equal to 1 each year, indicating high efficiency. The remaining cities were mostly in a state of moderate efficiency or inefficiency. Therefore, there was still significant room for improvement in AGTFP in some areas of Shandong province every year. In terms of regional distribution, the more efficient cities are mostly located in coastal areas and the central region. (2) From the perspective of dynamic analysis, Shandong province's AGTFP has been constantly improving. The cities with rapid growth include Binzhou, Zibo, Jinan, and Weihai, while the cities with slower growth include Zaozhuang, Laiwu, and Taian. There were significant regional differences in AGTFP growth. From the data in Table 8 it can be observed that the value of techch is generally greater than 1, while effch varies, with some values greater than 1 and some values less than 1. This indicates that the progress of agricultural technology level is the main driver of AGTFP improvement, a conclusion consistent with previous research by Kumar et al., 2008； and Sheng et al., 2020. On the other hand, the agricultural management level still needs to be enhanced. This result reminds us to pay attention to agricultural management by adopting advanced information technology, accelerating land transfer, and implementing other measures to further enhance AGTFP. (3) In terms of spatial disparity, the difference in AGTFP growth rates among cities had been increasing year by year. Based on the annual Malmquist index values, the 17 cities in Shandong province can be classified into three types: (1) Continuous growth, with an index greater than 1 every year, consisting of 8 cities; (2) Gradual growth, with an index transitioning from less than 1 to greater than 1, consisting of 3 cities; (3) Fluctuating growth, with an index mostly greater than 1 but occasionally less than 1, consisting of 6 cities. The results of the CoG model indicate that faster growth mainly occurred in the eastern and northern regions. This phenomenon leads us to pay attention to the development speed of AGTFP in the western and southern regions in order to achieve a more balanced AGTFP distribution. (4) Our research results indicate that the level of urbanization has no significant impact on AGTFP in Shandong province. This finding differs from the conclusion drawn by Li et al. (2021), who suggested a U-shaped relationship between urbanization and agricultural TFP in China. The discrepancy in research findings may be attributed to variations in the geographical scope of the studies. China encompasses over 30 provincial-level administrative regions with varying levels of development, which can lead to divergent conclusions due to differences in statistical data. In the case of Shandong province, the agricultural industrial structure was found to have an insignificant effect on AGTFP. This finding contrasts with previous research by Han et al. (2018) and Yang et al. (2019), who suggested that adjustments in agricultural industry structure have a negative impact on agricultural TFP. The reason for this difference is that the statistical data for Shandong province showed that the variation in grain crop sowing area was not significant during the study period, leading to the conclusion that the agricultural industry structure index had an insignificant impact on AGTFP. Similarly, financial support for agriculture in Shandong province was found to have an insignificant effect on AGTFP, which aligns with the conclusions of Liang and Xi (2022) and Liang and Long (2015) regarding the relationship between financial support for agriculture and agricultural TFP. This suggests that increasing financial support for agriculture may not be conducive to improving AGTFP (Wang et al., 2022). However, the industrialization of cities and the personal development of farmers were found to have a significant impact on AGTFP. The industrialization of cities was observed to have a 19 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 negative effect on AGTFP. This is attributed to the migration of a substantial number of young and middle-aged rural laborers from agriculture to the secondary and tertiary sectors as industrialization progresses, resulting in a shortage of labor in rural areas (Xu et al., 2022). Hence, this exodus of rural labor negatively impacts rural TFP. This conclusion aligns with the findings of Liang and Xi (2022) on AGTFP in Shandong Province, where they suggested that industrial development can create a certain siphoning effect on agricultural production factors, which is detrimental to the improvement of AGTFP. The conclusion that the personal development of farmers has a significant impact on AGTFP aligns with the findings of Paudel et al. (2004), who demonstrated a significant relationship between agricultural productivity and the quality of human capital across different states in the United States. Additionally, it corresponds to the results of Yang et al. (2022) in their study on the relationship between agricultural productivity and rural human capital in China. This suggests that improving the education level of farmers can contribute to the enhancement of AGTFP (Reimers and Klasen, 2013). The aforementioned research results underscore the regional variation in the factors influencing AGTFP. This aligns with the findings of Zhao et al. (2022) and Yang et al. (2019). To improve AGTFP in a specific region, it is imperative to conduct a thorough analysis of the local context and avoid adopting practices from other regions indiscriminately. The panel data regression results highlight the significance of mitigating the negative impact of industrialization and enhancing the quality of the labor force as crucial factors influencing AGTFP. 7 Conclusion and Suggestions 7.1 Conclusion In order to promote the sustainable development of agriculture, this paper studied the AGTFP of Shandong province. Through this study, we have discovered that the calculation of AGTFP with the inclusion of undesirable outputs yields lower results compared to calculations without considering undesirable outputs. We have found regional disparities in AGTFP within Shandong province, with approximately half of the regions consistently operating at medium to low efficiency levels each year. Over the study period, Shandong's AGTFP displayed an uninterrupted upward trajectory. The decomposition results of the Malmquist index indicate that the regional disparities in AGTFP within Shandong province were primarily influenced by the efficiency change component (effch), highlighting the need to improve management practices in the agricultural development process. By utilizing AGTFP gravity calculations, we observed spatial variations in AGTFP efficiency, with the AGTFP gravity center shifting towards the east and north. In our panel regression analysis, we found that industrialization and the personal development of farmers have a significant impact on AGTFP. Therefore, there are opportunities to mitigate rural labor outmigration by enhancing rural public services, thereby improving the living conditions of farmers through better healthcare, education, transportation, and other essential amenities. Additionally, the development of the agro-processing industry and the implementation of rural tourism initiatives can help in this regard. Enhancing the personal development of farmers through technical training and educational programs can also contribute to improving AGTFP. The findings of this research can provide valuable insights for the agricultural green development in Shandong and China as a whole. Due to limitations in data availability, our analysis of factors influencing AGTFP may not be exhaustive. In the future, we will continue to collect data and delve deeper into the exploration of factors affecting AGTFP, providing more informed recommendations for its improvement. 20 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 7.2 Suggestions In order to improve Shandong AGTFP and promote regional sustainable development and to balance the province-wide development, this paper proposes: (1) Improve agricultural management and efficiency and reduce extensive management. Reducing factor inputs, especially fertilizers and pesticides, can not only improve efficiency, but also reduce pollution. Advanced technology should be adopted for farmland irrigation to reduce waste. (2) Focus on industrial structure transformation, improve the efficiency of the primary industry and reduce pollution, and emphasize on developing projects with low pollution emissions and high efficiency, such as agricultural sightseeing tourism and ecological agriculture. (3) Improve the personal development of farmers. By improving the level of agricultural technology and management through the enhancement of farmers' personal development, the overall efficiency and productivity of agricultural operations can be significantly enhanced, ultimately leading to improved AGTFP. Acknowledgements Thanks to the peer reviewers and journal staff for their valuable suggestions and diligence work in improving the paper for publication. References Alhassan, H. (2021): The effect of agricultural total factor productivity on environmental degradation in sub-Saharan Africa. Scientific African 12: e00740. https://doi.org/10.1016/j.sciaf.2021.e00740. Bai, Z. (2008): Econometric Analysis of Panel Data. Nankai University Press, Tianjin. Caves D.W., Christensen, L.R., Diewert, W.E. (1982): The economic theory of index numbers and the measurement of input, output, and productivity. Econometrica 50 (6): 1393-1414. https://doi.org/10.2307/1913388. Chinairn (2020): China is a major producer and consumer of agricultural and sideline products, what is the current situation of the agricultural product processing industry?. https://www.chinairn.com/hyzx/20201224/143356816.shtml. Cnr (2014): China's excess capacity in chemical fertilizers needs to be addressed. http://country.cnr.cn/tjxw/201402/t20140207_514794889.shtml. Espoir, D.K., Bannor, F., Sunge, R. (2021): Intra-Africa agricultural trade, governance quality and agricultural total factor productivity: Evidence from a panel vector autoregressive model. ZBWLeibniz Information Centre for Economics, Kiel, Hamburg. https://hdl.handle.net/10419/235617. Fang, L., Hu, R., Mao, H. Chen, S.J. (2021): How crop insurance influences agricultural green total factor productivity: Evidence from Chinese farmers. Journal of Cleaner Production, vol. 321: 128977. https://doi.org/10.1016/j.jclepro.2021.128977. Fuglie, K. (2015): Accounting for growth in global agriculture. Bio-based and applied economics 4 (3): 201-234. https://doi.org/10.13128/BAE-17151. Gao, T.M., Wang, J.M., Chen, F., Liu, Y.H. (2016): Econometric Analysis Methods and Modeling: EViews Applications and Examples. Tsinghua University Press, Beijing. http://www.tup.tsinghua.edu.cn/booksCenter/book_06018001.html. 21 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 Han, H., Zhong, Z., Wen, C., Sun, H. (2018): Agricultural environmental total factor productivity in China under technological heterogeneity: characteristics and determinants. Environmental Science and Pollution Research 25: 32096-32111. https://doi.org/10.1007/s11356-0183142-4. Hong, M.Y., Tian, M.J., Wang, J. (2022): Digital Inclusive Finance, agricultural industrial structure optimization and agricultural green total factor productivity. Sustainability 14 (18): 11450. https://doi.org/10.3390/su141811450 Hsiao, C. (2007): Panel data analysis-advantages and challenges. Test 16 (1):1-22. https://doi.org/10.1007/s11749-007-0046-x. Huang, Y.L., Zhang, L., Zhu, C.X. (2012): Analysis of agricultural non-point source pollution in nansi lake basin, Shandong Province. Environmental Science Research 25 (11): 12431249. https://doi.org/10.13198/j.res.2012.11.57.huangyl.001. Jiao, Y. (2013): Evaluation of agricultural production efficiency in Shandong province. China Population·Resources and Environment 23 (12):105-110. https://doi.org/10.3969/j.issn.1002－2104.2013.12.016. Jingweidu (2020): Longitude and Latitude Query. https://jingweidu.51240.com/. Lai, S.Y. (2003): Non-point source survey and evaluation method and its application research. Tsinghua University, Beijing. Lai, S.Y., Du, P.F., Chen, J.N. (2004): Non-point source pollution investigation and evaluation method based on unit analysis. Journal of Tsinghua University (Natural Science Edition) 44 (9): 1184-1187. https://doi.org/10.16511/j.cnki.qhdxxb.2004.09.009. Li, J., Chen, J., Liu, H. (2021): Sustainable agricultural total factor productivity and its spatial relationship with urbanization in China. Sustainability 13 (12): 6773. https://doi.org/10.3390/su13126773. Li, M. (2019): Research on the shift of the center of gravity and spatial coordination of “population-economy-land” in Shandong cities. Shandong Normal University, Jinan. Li, Z.D., Yuchi, X.J. (2019): Research on agricultural production efficiency in Shandong province. Journal of Shandong Agricultural University (Social Science Edition), 21(02): 45-51. Li, Y., Chen, C., Liu, F., Wang, J. (2022): Analysis method of agricultural total factor productivity based on stochastic block model (SBM) and machine learning. Journal of Food Quality, vol. 2022. https://doi.org/10.1155/2022/9297205. Liang, J., Long, S.B. (2015): Green agricultural total factor productivity growth and its influencing factors. Journal of South China Agricultural University (Social Science Edition) 14 (3): 1-12. https://doi.org/10.7671/j.issn.1672-0202. 2015.03.001. Liang, J.R., Xi, X.R. (2022): Measurement of Agricultural Green Total Factor Productivity and Analysis of Influencing Factors in Shandong Province. Rural Economy and Technology 33 (5): 1-5+15. Liang, L.T. (2009): Research on the spatio-temporal characteristics and evolution of rural ecological environment. Nanjing Agricultural University, Nanjing. Liu, D.D., Zhu, X.Y., Wang, Y.F. (2021): China's agricultural green total factor productivity based on carbon emission: An analysis of evolution trend and influencing factors. Journal of Cleaner Production 278 123692: 1-12. https://doi.org/10.1016/j.jclepro.2020.123692. Liu, F., Zhang, H.L. (2018): Analysis of Agricultural Total Factor Productivity and Influencing Factors under Carbon Emission Constraints in Shandong Province. Jiangsu Agricultural Science 46 (24): 387-390. https://doi.org/10.15889/j.issn.1002-1302.2018.24.099. Liu, J.X., Wang, M.J., Yang, L., Rahman, S., Sriboonchitta, S. (2020): Agricultural productivity growth and its determinants in south and southeast Asian countries. Sustainability 12 (12): 4981. https://doi.org/10.3390/su12124981. Liu, Z.W. (2018): Analysis on the dynamic evolution and influencing factors of agricultural total factor productivity in china. Chinese Journal of Agricultural Resources and Regional Planning 39 (12): 104-111. https://doi.org/10.7621/cjarrp.1005-9121.20181215. 22 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 Kumar, P., Mittal, S., Hossain, M. (2008): Agricultural Growth Accounting and Total Factor Productivity in South Asia: A Review and Policy Implications. Agricultural Economics Research Review 21: 145-172. Ma, J.Z., Ding, Z.Y., Wei, G.X., Zhao, H., Huang, T.M. (2009): Sources of water pollution and evolution of water quality in the Wuwei basin of Shiyang River，Northwest China. Journal of Environmental Management 90 (2): 1168-1177. https://doi.org/10.1016/j.jenvman.2008.05.007. National Bureau of Statistics of the People's Republic of China (2020): China Statistical Yearbook, 2009-2019. http://www.stats.gov.cn/. Paudel, K.P., Sambidi, P.R., Sulgham, A.K. (2004): A Theoretical Development and Empirical Test on the Convergence of Agricultural Productivity in the USA. Conference Paper. Presentation at AAEA at 2004 annual meeting. https://doi.org/10.22004/ag.econ.20175. Rmlt (2019): The economic achievements of the 40 years of reform and opening up and the "business card of China". http://www.rmlt.com.cn/2019/0520/547309.shtml. Reimers, M., Klasen, S. (2013): Revisiting the role of education for agricultural productivity. American Journal of Agricultural Economic 95 (1): 131-152. https://doi.org/10.1093/ajae/aas118. Sannong (2014): Looking forward to solving the overcapacity of fertilizer production. http://sannong.cntv.cn/20140207/101694.shtml. Sathye, M. (2002): Measuring productivity changes in Australian banking: an application of Malmquist indices. Managerial Finance 28 (9): 48-59. https://doi.org/10.1108/03074350210768068. Shandong Provincial Bureau of Statistics (2020): Shandong Statistical Yearbook (2009-2019). http://tjj.shandong.gov.cn/col/col6279/index.html. Sheng, Y., Tian, X., Qiao, W., Peng, C. (2020): Measuring agricultural total factor productivity in China: pattern and drivers over the period of 1978-2016. Australian Journal of Agricultural and Resource Economics 64 (1):82-103. https://doi.org/10.1111/1467-8489.12327. Sheng, Y., Ball, V.E., Erickson, K., Mesonada, C.S.J. (2022): Cross-country agricultural TFP convergence and capital deepening: Evidence for induced innovation from 17 OECD countries. Journal of Productivity Analysis 58 (2-3): 185-202. https://doi.org/10.1007/s11123022-00646-z. Song, T.S., Zhang, Z.T., Hao, T. (2012): A study on agricultural production's efficiency during the 11th Five-year Plan period in Shandong Province on basis of DEA method. Journal of Qilu Normal University 27 (3): 25-31. Tan, R.H., Liu, H.M. (2022): Characteristic evolution and influencing factors of the spatial correlation network of agricultural green total factor productivity in China. Chinese Journal of Eco-Agriculture 30 (12): 2011-2026. https://doi.org/10.12357/cjea.20220339. Tellier, L.N., Vertefeuille, C. (1995): Understanding spatial inertia: center of gravity, population densities, the Weber problem, and gravity potential. Journal of Regional Science 35 (1): 155-164. https://doi.org/10.1111/j.1467-9787.1995.tb01404.x. Tone, K. (2001): A slacks-based measure of efficiency in data envelopment analysis. European Journal of Operational Research 130 (3): 498-509. https://doi.org/10.1016/S03772217(99)00407-5. Tu, Z.G., Liu, L.K. (2011): Evaluation of China’s industrial efficiency considering energy and environmental factor. Economic Review (2): 55-65. https://doi.org/10.19361/j.er.2011.02.007. USDA (2012): Total factor productivity has become the primary source of growth in world agriculture. https://www.ers.usda.gov/dataproducts/chart-gallery/gallery/chart-detail/?chartId=76219. Wang, L., Tang, J.Y., Tang, M.Q., Su, M.Y., Guo, L.L. (2022): Scale of operation, financial support, and agricultural green total factor productivity: Evidence from China. International 23 Peng et al. | Ger J Agr Econ 73 (2024), No. 2 Journal of Environmental Research and Public Health 19 (15): 9043. https://doi.org/10.3390/ijerph19159043. Wang, S., Wang, X. (2017): Does service inward foreign direct investment improve green total factor productivity: An empirical study based on china’s provincial panel data. Journal of International Trade issue 12: 83-93. https://doi.org/10.13510/j.cnki.jit.2017.12.008. Wang, Y.R., Xie L.Y. (2022): Research on the Impact of Human Capital on Agricultural Green Total Factor Productivity: An Analysis of the Impact Path Based on Mediation Effect. Journal of Shandong Agricultural University (Social Science Edition) 24 (3): 1-7. https://mall.cnki.net/magazine/article/SDND202203001.htm. Wang Y.F., Xiao, Y. (2022): New Urbanization, Agricultural Industry Agglomeration, and Agricultural Green Total Factor Productivity Growth: Impact Mechanism and Empirical Testing. Journal of Chongqing Normal University (Social Sciences Edition) 42 (4): 5-20. https://doi.org/10.19742/j.cnki.50-1164/C.220401. Xu, Q.H., Zhu, P.X., Tang, L. (2022): Agricultural services: Another way of farmland utilization and its effect on agricultural green total factor productivity in China. Land 11 (8): 1170. https://doi.org/10.3390/land11081170. Xu, X.C., Zhang, L., Chen, L.H., Liu, C.J. (2020): The Role of soil N2O emissions in agricultural green total factor productivity: An empirical study from china around 2006 when agricultural tax was abolished. Agriculture 10 (5): 150. https://doi.org/10.3390/agriculture10050150. Yang, J., Wang, J., Li, C., Liu, X.P. (2019): Spatial differentiation and driving factors of green total factor productivity in China's agriculture. Quantitative Economic and Technological Economic Research 36 (10): 21-37. https://doi.org/10.13653/j.cnki.jqte.2019.10.002. Yang, Y., Ma, H., Wu, G.S. (2022): Agricultural green total factor productivity under the distortion of the factor market in China. Sustainability 14 (15): 9309. https://doi.org/10.3390/su14159309. Ye, F., Yang, Z.N., Yu, M., Watson, S., Lovell, A. (2023): Can Market-Oriented Reform of Agricultural Subsidies Promote the Growth of Agricultural Green Total Factor Productivity? Empirical Evidence from Maize in China. Agriculture 13 (2): 251. https://doi.org/10.3390/agriculture13020251. Zhang, H.Y., Liu, Y.Z. (2015): Evaluation and analysis of agricultural production efficiency in Shandong Province based on undesirable output model. Journal of Henan Agricultural University 49 (3): 423-428. https://doi.org/10.16445/j.cnki.1000-2340.2015.03.024. Zhang, P.Y., Li, Y.Y., …, Qin, M.Z. (2020a): Comprehensive assessment of the effect of urban built-up land expansion and climate change on net primary productivity. Complexity, 2020:112. https://doi.org/10.1155/2020/8489025. Zhang, P.Y., Yang, D., …, Shao, J.W. (2020b): Re-examining the drive forces of China’s industrial wastewater pollution based on GWR model at provincial level. Journal of Cleaner Production vol. 262: 121309. https://doi.org/10.1016/j.jclepro.2020.121309. Zhao, P.Y., Wu, H.Y., Lu, Z.G., Kou, J.L. and Du, J. (2022): Spatial differences, distributional dynamics, and driving factors of green total factor productivity in China. Frontiers in Environmental Science 10: 1058612. https://doi.org/10.3389/fenvs.2022.1058612. Zuo, Y.X. (2019): Study on the difference of production and scale of agricultural total factor productivity based on TFP Index. Ekoloji 28 (107): 3729-3734. Contact Author Dr. Jay Lee Department of Geography, Kent State University Kent, 44240, OH, USA e-mail: [email protected] 24