Research on the Development of Agricultural Complex in Chinese Provinces Based on Input ‐ Output Analysis

: Chinese provinces and regions have huge differences in agricultural resource endowments and uneven levels of agricultural industry development, resulting in inconsistent influence of agricultural complexes on the economic development of China's provinces and regions. This paper uses the input-output model to divide the agricultural complex into four links: agricultural production input, production, agricultural product processing and final distribution of agricultural products. By estimating the GDP of agricultural complex in 30 provinces and cities in China in 2002, 2007, 2012 and 2017, the paper analyzes the structure composition and change trend of the agricultural complex's share of GDP and the four links of agricultural complex. On this basis, this paper uses the two indicators of agricultural resource abundance and agricultural product processing development to divide the agricultural complexes in 30 provinces and regions of China into four types: the first type is the area with rich agricultural resources and developed processing industry. The second category is the area with developed agricultural resources and backward processing industry. The third category is the area with poor agricultural resources and developed processing industry. The fourth category is the region with poor agricultural resources and backward processing industry, which provides a reference basis for the coordinated development of agricultural industries in various provinces and regions and complements each other's advantages.


Introduction
With the revolutionary development of agricultural technology after World War II, agriculture has undergone a transformation from traditional self-sufficient agriculture to a generalized and specialized agricultural production model.Studying the methods of agricultural production, processing, and distribution, as well as the input factors and production functions of agricultural production, has become a new topic in the academic community.
In the process of agricultural development, traditional industry classification has revealed its limitations.This classification method divides different economic activities into independent primary, secondary, and tertiary industries, ignoring the complex relationships that exist in input, output, processing, and distribution.To overcome this limitation, Goldberg (1968) proposed the concept of "agricultural complexes", which stems from the complex relationships that exist in the input, output, processing, and distribution of agricultural production.
Furtuoso & Guilhoto (1998) proposed a basic method for estimating the GDP of agricultural complexes, which divides them into four parts: agricultural input, agricultural production, agricultural product processing industry, and final distribution.Van Leeuwen (2000) divided agricultural complexes into four parts: agricultural inputs, agricultural production, agricultural product processing inputs, and agricultural product processing inputs.He calculated the proportion of agricultural complexes GDP in major European countries to their respective GDP, with Germany, the United Kingdom, France, and Italy all having a proportion ranging from 5.1% to 8.5%.Similarly, Furtuoso & Guilhoto (2003) used the Brazilian input-output table to measure the internal structure of Brazilian agricultural complexes and their proportion to GDP.The calculation showed that in 2000, the proportion of Brazilian agricultural complexes to Brazil's GDP was 27%.
In China, literature on agricultural complexes mainly focuses on the following three aspects: firstly, studying agricultural complexes from both strategic and macro perspectives (Zhao Xia and Wu Fangwei, 2008; Geng Xianhui, 2009); Secondly, study the micro economic behaviors of small farmers, family farms, farmer cooperatives, and agricultural related enterprises (Tang Wenhua, 2019); Thirdly, using input-output models to study the relationship between agriculture and its related industries (Xue Jiliang, Li Lutang, 2011; Geng Xianhui, Zhou Yingheng, 2011).
After studying existing literature in China, the author found that there are currently shortcomings in using input-output models to study agricultural complexes.Firstly, using the China input-output table, a comparative analysis was conducted on the degree of linkages and industrial impact of China's agricultural industries.However, only forward and backward related industries were calculated, and there was insufficient research on the scale and structure of agricultural complexes.Secondly, in terms of spatial dimensions, comparisons are made between China's agricultural complexes and several developed countries with significant differences in agricultural resource endowments, such as the United States, Japan, and the Netherlands.However, there is insufficient comparison between domestic provinces (Yan Binjian, Fan Jin, Zhou Yingheng, 2011; Liang Yinghui, Jiang Zhihua, 2019).
The generalization and specialization of agriculture are occurring in various regions around the world, depending on the stage of economic and social development in each region.Among them, the degree of participation of agricultural institutions (such as farmers, providers of agricultural inputs, processors of agricultural products, and distributors of agricultural products) in economic activities will show diversity.China has a vast territory, with significant differences in agricultural resource endowments and uneven levels of agricultural industry development in different regions.Studying the GDP and internal structure of agricultural complexes in various provinces and regions of China is of practical significance for the coordinated development, complementary advantages, and enhancement of agricultural industries in each province and region.
In the second part, this paper introduces the method of using the input-output model to measure the GDP of agricultural complexes.In the third part, it estimates the GDP of modern agricultural complexes in 30 provinces and regions of China (excluding Xizang) in 2002, 2007, 2012 and 2017.At the same time, it compares and analyzes the internal structure of agricultural complexes in various provinces and regions, and on this basis, it divides agricultural complexes in various provinces and regions of China into four types.Finally, provide analytical conclusions in the fourth section.

Research Method
This section adopts the measurement method for agricultural complexes proposed by Zhao Xia and Wu Fangwei (2008), which is applicable to the regional inputoutput table in China.This method divides the GDP estimation of regional agricultural complexes in China into four stages: agricultural production input, agricultural production, agricultural product processing and manufacturing, and final distribution of agricultural products.

GDP of agricultural production input link
Use the value of agricultural inputs in the input-output table to estimate the GDP of agricultural production inputs: Multiply the sequence vectors of input value and value-added coefficient in the agricultural sector.The agricultural input value sequence is directly taken from the input-output table.The specific calculation formula is as follows: In the formula, GDPl represents the GDP of the input; ik Z is the consumption of the i-th product in the agricultural sector; i CAV is the added value coefficient of sector i.

Agricultural production GDP
The estimation of GDP in the agricultural production process is the estimation of the added value of the agricultural sector.Considering that the estimation of GDP in the input process includes the added value of products or services as intermediate inputs in the sector, this duplicate value needs to be eliminated.Therefore, the estimated GDP of the production process is as follows: In the formula:

GDP of agricultural product processing and manufacturing
Utilize the connections between departments to determine the added value of agricultural product processing and manufacturing.The agricultural product processing and manufacturing process mainly includes the following five departments: ① food and tobacco processing industry; ② Textile industry; ③ Wood furniture manufacturing industry; ④ Clothing products industry; ⑤ Paper printing and stationery manufacturing industry.
When estimating the GDP of agricultural product processing and manufacturing, it is also necessary to eliminate the double calculation impact, that is, the added value of the processing and manufacturing sector minus the value of the corresponding sector's input.The calculation formula for GDP in this stage is as follows: ) ( Among them, III GDP is the GDP of agricultural product processing and manufacturing; q is the relevant sector in the agricultural product processing and manufacturing process; q CVA is the added Value coefficient of agricultural product processing and manufacturing industry.

GDP of final distribution of agricultural products
The final distribution link GDP of agricultural products mainly considers the added value of agricultural sectors in transportation, commerce, and other service sectors.Assume that the added value of this part is positively correlated with the added value of the service industry in the economy, as well as with the final demand for agriculture and agricultural product processing industry.The specific calculation formula is: Among them, TM is the total added value of the circulation sector, commercial trade sector, and other service sectors, that is, the added value of the service industry; TFD is the overall final demand; k FD is the final demand in the agricultural production process; q FD is the final demand of the q-th sector in the agricultural product processing and manufacturing process; IV GDP is the GDP of the final distribution process of agricultural products.The final demand mentioned above includes consumption, investment, and net exports.In summary, the expression for the GDP of agricultural complexes is: In the formula: The data used in this article comes from the input-output tables of 42 secors in various regions in 2002, 2007, 2012, and 2017.

Empirical Results
This section first measures the GDP of the four links of agricultural complexes in various provinces and regions of China in 2017; Then, use the time-series regional input-output table to calculate the development status of agricultural complexes in different provinces and regions of China in different periods, and measure the correlation between the GDP of the four links of agricultural complexes; Finally, based on the accounting results of agricultural complexes, classify the development of agricultural complexes in various provinces and regions of China.

The total GDP and its composition of agricultural comprehensive bodies in each province and region
The estimated GDP results of the four links of agricultural complexes in various provinces and regions of China in 2017 are shown in Table 1.For the 30 provinces and regions in China, the total GDP of agricultural complexes is 19899.89 billion yuan.Among them, the GDP of input, production, processing, and final distribution were 1808.71 billion yuan, 5693.04 billion yuan, 6516.49 billion yuan, and 5881.64 billion yuan, respectively, with a relative structure of 9.09%, 28.61%, 32.75%, and 29.56%, respectively.The processing and final distribution of agricultural products have occupied the dominant position in agricultural complexes.
In 2017, the top three provinces in terms of GDP for agricultural complexes are Guangdong with a total of 1805 Data source: Calculated based on the regional input-output table of 30 provinces and cities in China in 2017 The GDP of agricultural production in 2017 is 5693.04 billion yuan, while the GDP of agricultural complexes is 19899.89 billion yuan, which is 3.50 times that of the former.Excluding Beijing (due to the special nature of administrative functions), there is a significant difference in the ratio between different provinces and regions.The wealthier the province, the higher the ratio.

The proportion of agricultural complexes in GDP in each province and region
The GDP of China's agricultural complexes is more than three times that of agricultural production links, and there is a trend of increase.In 2002, the GDP of agricultural production was 1404.60 billion yuan, while the GDP of agricultural complexes was 4413.22 billion yuan, which is 3.14 times that of the former; In 2007, the GDP of agricultural production was 2392.81 billion yuan, while the GDP of agricultural complexes was 8363.42 billion yuan, which is 3.49 times that of the former; In 2012, the GDP of agricultural production was 4475.87 billion yuan, while the GDP of agricultural complexes was 16476.5 billion yuan, which is 3.68 times that of the former; The GDP of agricultural production in 2017 is 5693.04 billion yuan, while the GDP of agricultural complexes is 19899.89 billion yuan, which is 3.50 times that of the former.As shown in Table 2, the proportion of agricultural complexes to GDP in the vast majority of provinces is showing a downward trend.Out of 30 provinces and cities, 29 have experienced a decline, with only Heilongjiang experiencing a slight increase.The largest decrease was in Inner Mongolia, which decreased by 31.99 percentage points.Secondly, Yunnan decreased by 27.88 percentage points, Anhui decreased by 23.24 percentage points, and Shandong decreased by 23.36 percentage points.There are 13 provinces and cities with a decline range of 10% to 20%, namely Jilin, Jiangsu, Guangxi, Sichuan, Zhejiang, Hunan, Jiangxi, Hebei, Tianjin, Hubei, Chongqing, Beijing, Shanxi, Shanghai, and Liaoning.Although the decline in these provinces is relatively small, they account for the majority of all provinces in the country.The other 9 provinces and cities have decreased by less than 10%, namely Guangdong, Qinghai, Guizhou, Henan, Ningxia, Shaanxi, Gansu, Xinjiang, and Heilongjiang.
The reasons for the changing trends in various provinces and regions are different: for example, the proportion of GDP in agricultural input, production, and distribution in Heilongjiang has increased, although the proportion of processing has decreased, overall it still rises; For Fujian, the main reason is that the proportion of GDP in the processing sector has increased, although the proportion of input, production, and distribution sectors has decreased, overall it still rises; However, the proportion of GDP in agricultural input, production, processing, and distribution in Inner Mongolia has decreased, and overall, it has naturally declined.
Table 3 shows the ratio of input, production, processing, and distribution in agricultural complexes to GDP.On average, in 2002, the ratio was 0.048, 0.138, 0.079, and 0.085, indicating an advantage in the production process.By 2017, the ratios will be 0.021, 0.067, 0.077, and 0.070 respectively, with the processing sector accounting for the highest proportion of GDP, followed by the final distribution sector.The production sector is relatively less important.Based on Table 1, it is not difficult to see that in the process of optimizing industrial structure, the proportion of agricultural complexes in GDP has decreased in each province and region, but the deep processing and distribution services of agricultural products continue to develop and grow.

The correlation between the four link GDP of agricultural complexes in various provinces and regions
The four stages of input, production, processing, and final distribution are relatively independent, but also closely related.Its independence is reflected in the different divisions of economic activities, and its closeness is reflected not only in the chronological order of the four links, but also in the interdependence between input and output in each link.
Calculating the correlation between the four stages of GDP has two purposes: firstly, indicators with strong correlation region, dividing them into four categories (see Table 5 for details).Category I: Rich agricultural resources and developed processing industry; Category II: Rich agricultural resources and backward processing industry; Category III: Poor agricultural resources and developed processing industry; Category IV: Poor agricultural resources and backward processing industry.No province belongs to Class I region; The regions belonging to Class II include Hebei, Shanxi, Inner Mongolia, Liaoning, Heilongjiang, Anhui, Hubei, Guangxi, Hainan, Chongqing, Sichuan, Guizhou, Yunnan, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang; The regions belonging to Class III include Beijing, Tianjin, Jilin, Shanghai, Jiangsu, Zhejiang, Fujian, Jiangxi, Shandong, Henan, and Guangdong; Hunan belongs to the fourth category of provinces and cities.

Conclusion
The empirical results can infer the complexity of agricultural complexes.This article uses the input-output tables of various provinces and cities in China for 2002, 2007, 2012, and 2017 to calculate.It is found that the GDP of agricultural complexes is significantly higher than that of agriculture, with the former being 3.14, 3.49, 3.68, and 3.50 times that of the latter.The ratio varies greatly among different provinces and regions, and the wealthier the province, the higher the ratio.This also indicates to some extent that the development direction of agricultural complexes is constantly extending towards areas such as deep processing of agricultural products, convenient logistics, and thoughtful services.
Agricultural complexes have played an important role in the Chinese economy.For China in 2017, the proportion of agricultural GDP (production) is 6.73%, while the proportion of agricultural complex GDP is 23.52%.When the agricultural industry takes agricultural complexes as the research object, its multiplier effect is amplified.When examining agricultural complexes in various provinces and regions, the proportion of agricultural GDP ranged from 0.31% to 20.84%, and the proportion of agricultural complex GDP ranged from 7.11% to 39.73%.There are significant differences among different provinces and regions.
In the four links of agricultural complexes, the share of agricultural product processing and final distribution continues to expand, while the proportion of input and production decreases.In 2002, the structural coefficients of the four links were 11.58%, 31.83%,27.28%, and 29.31%, respectively; By 2017, the structural coefficients will be 9.09%, 28.61%, 32.75%, and 29.56%, respectively.In the process of optimizing industrial structure in various provinces and regions, the proportion of agricultural complexes in GDP has decreased, but the deep processing and distribution services of agricultural products continue to develop and grow.
There are differences in the correlation between the four links of agricultural complexes.The input and production processes are highly correlated, while the processing and final distribution processes are highly correlated.This article calculates the two indicators of agricultural resource abundance and processing development, and divides agricultural complexes in 30 provinces and regions of China into four types based on the relative sizes of the two indicators.This can provide reference for the coordinated development and complementary advantages of agricultural industries in each province and region.
II GDPrepresents the GDP of agricultural production, k VA represents agricultural added value, k CVA is the coefficient of agricultural added value, and kk Z is the intermediate input of agricultural products itself.

Table 1 .
.32 billion yuan, Shandong with 1743.51 billion yuan, and Jiangsu with 1639.00 billion yuan.The fourth to thirteenth provinces are Henan with 1432.65 billion yuan, Zhejiang with 1278.86 billion yuan, Fujian with 1164.67 billion yuan, Hubei with 1061.14 billion yuan, Sichuan with 1047.46 billion yuan, Hunan with 967.26 billion yuan, Hebei with 898.71 billion yuan, Anhui with 752.43 billion yuan, Guangxi with 677.77 billion yuan, and Heilongjiang with 631.85 billion yuan.The provinces with high GDP rankings in agricultural production include Shandong Province with a GDP of 468.02 billion yuan, Jiangsu Province with a GDP of 378.14 billion yuan, Sichuan Province with a GDP of 378.01 billion yuan, Henan Province with a GDP of 375.93 billion yuan, Guangdong Province with a GDP of 335.31 billion yuan, Hubei Province with a GDP of 330.45 billion yuan, Hebei Province with a GDP of 299.64 billion yuan, Hunan Province with a GDP of 274.80 billion yuan, Guangxi with a GDP of 256.26 billion yuan, Heilongjiang Province with a GDP of 245.28 billion yuan, Anhui Province with a GDP of 233.59 billion yuan, Fujian Province with a GDP of 208.28 billion yuan, Yunnan Province with a GDP.Summary of GDP estimates for the four links of agricultural complexes in 2017 Unit: 100 million yuan Provinces greater than 4 include Shanghai 23.16, Tianjin 9.44, Zhejiang 6.42, Fujian 5.59, Guangdong 5.38, and Jiangsu 4.33; The provinces between 3 and 4 are Henan 3.8, Jilin 3.81, Shandong 3.73, Jiangxi 3.56, Hunan 3.52, Anhui 3.22, Hubei 3.21, and Hebei 3.00; The provinces between 1 and 2 are Shaan 2.96, Yun 2.80, Ning 2.79, Chuan 2.77, Gui 2.74, Yu 2.69, Gui 2.64, Hei 2.58, Gan 2.38, Xin 2.28, Jin 2.25, Liao 2.22, Inner Mongolia 2.21, Qing 2.02, and Qiong 1.85.This also indicates to some extent that the development direction of agricultural complexes is constantly extending towards areas such as deep processing of agricultural products, convenient logistics, and thoughtful services.

Table 2 .
Summary of the proportion of agricultural complexes to GDP

Table 5 .
Summary of Agricultural Resource Abundance and Processing Development in Various Provinces and Regions of Calculated based on the regional input-output table of various provinces and cities in China in 2022