Research on the Influence of Urbanization Structure on Farmers’ Income Increase—Based on the Perspective of Income Structure

Different from the existing studies that examine the impact of the overall development level of urbanization on farmers’ income, this article empirically studies the impact of two aspects of urbanization level structure, “citilization” and “townization” on farmers’ income. Based on the empirical study of 31 provincial panel data from 2007 to 2018, it is found that “citilization” and “townization” can significantly improve farmers’ income, and the role of “townization” is obviously greater than that of “urbanization”. In terms of the impact of income components, “citilization” and “townization” have no obvious impact on wage income and operating income, but promote both property income and transfer income, and “townization” has an even greater effect. The policy implication of the research conclusion is that while using urban “engine” to drive economic growth, we should actively use the pivotal role of small towns to connect urban and rural development to promote rural re-vitalization and development.

party and the state must solve at the moment.
The academic research on the factors affecting farmers' income is mostly based on economic growth theory, mainly from the following three perspectives: first, from the perspective of capital, some scholars believe that financial support for agriculture can increase farmers' transferable income, then directly and indirectly affect the level of farmers' income. For example, Xiao & Xu (2019) analyzed that fiscal support agriculture policies could not only significantly increase farmers' income, but also effectively guarantee and regulate farmers' income; Zhu & Lu (2020) studied the effects of agricultural subsidy policy on supporting agriculture through the method of income mobility matrix and panel quantile regression, and concluded that financial subsidies could help alleviate absolute poverty in rural areas and increase the income of low-income groups. Second, some scholars study from the perspective of technology. For example, Wang & Hu (2019) constructed a model from a theoretical aspect, and found empirically that agricultural productivity mainly increased farmers' income by increasing farmers' operational income; Ma & Kong (2019) studied the relationship between agricultural technological progress and the income gap of farmers between regions, and concluded that agricultural technological progress had a positive effect on farmers' income. Third, scholars' research from the perspective of labor mainly focuses on the education level of farmers and labor transfer. For example, Song (2010) used the national provincial panel data from 1985 to 2005 to conduct empirical research. He believed that education level has an increasing effect on farmers' income, and the marginal effect of education on income had gradually increased; Li, Li, & Zhou (2018) constructed a dynamic panel data model, and the analysis showed that labor transfer could significantly promote the growth of farmers' total income and non-agricultural income; Liu & Pan (2019) used the PVAR model to empirically showed that the impact of labor transfer on farmers' income would increase with the increase of the number of labor transfers.
In the current research on the relationship between urbanization and farmers' income, most scholars explore the impact of the overall level of urbanization on farmers' income, mainly using the following two methods to demonstrate: first, using time series data models, such as Fan & Wang (2007) using the vector error correction model and pre variance decomposition, the study found that there was a positive long-term and stable co-integration relationship between urbanization development and farmers' income increase in China; Zhang, Huang,, & Li (2014) used vector autoregressive models and other methods to analyze the problem, then found there was a positive interaction between the process of urbanization and the increase in farmers' income. Second, using provincial panel data models, such as Wang & Peng (2013) using 1999-2011  to build a variable intercept model, the empirical evidence showed that the increase in urbanization could effectively increase farmers' income; Wang & Zhu (2014) used the Panel Data model and found that the promotion of urbanization had a positive correlation with the increase of farmers' income. In addition, some scholars have studied the impact of urbanization development on the components of farmers' income. For example, Ye, Xu, & Hu (2010) used the grey relational theory analysis to find that the impact of urbanization on farmers' income sources was as follows: wage income, transfer income, property income, operating income; Yuan & Zhou (2015) researched that the effect of urbanization on the various components of farmers' income was closely related to the level of regional urbanization development.
The above studies have provided many methods for discussing the impact of the overall development of urbanization on farmers' income. The research results are quite rich, but few scholars have studied the impact of urbanization structure on the increase of farmers' income. At present, the main trend of China's urbanization development is reflected in the "citilization" and "townization" of population gathering. In the future, the population growth of town areas will exceed the population growth of urban areas. The urbanization of town areas is the main driving force for the rapid development of China's urbanization.
Therefore, in the process of urbanization in a region, is "citilization" having a greater impact on farmers' income, or is "townization" more able to promote farmers' income? Based on this, this article uses 31 provincial-level panel data to deeply explore the impact of urbanization structure on farmers' income and farmers' income components from the two aspects of "citilization" and "townization", and interpret its impact, so as to broaden the research fields of urbanization and farmers' income.

Urbanization Structure
According to the usual definition, this article uses the ratio of urban population to total population as the urbanization rate, the ratio of city population to total population as the citilization rate, and the ratio of town population to total population as the townization rate. Analyze and process the sample survey data in China Population and Employment Statistical Yearbook from 2003 to 2018 to obtain the urbanization rate, citilization rate, and townization rate. The results are shown in Table 1, and the results are drawn into a line chart ( Figure 1).
Then calculate the proportion of citilization rate in urbanization rate and the proportion of townization rate in urbanization rate in each year, and draw the change trend of these two indicators into Figure 2.
As can be seen from Figure 1 and    Correspondingly, it can be seen from Figure

Farmers' Income Structure
Obtain  Table 2.
It can be seen from Table 1 that the annual growth rate of the total income of farmers and its components at each stage is more than 6%, and the income of The average annual growth rate of farmers' income components is sorted by size: transfer income, property income, wage income, and operating income. Specifically, the average annual growth rate of farmers' wage income and the average annual growth rate of operating income have basically the same trend. Both showed an upward trend during 2004-2013, and the growth rate declined during 2015-2018. The possible reason is that between 2004 and 2013, the household registration system was gradually liberalized, and rural surplus labor was transferred from the agricultural sector to the non-agricultural sector on a large scale. The farmers' wage income had increased rapidly, and gradually became the main driving force to increase the income of farmers. Meanwhile, the rapid economic development in this period had also led to the increase of farmers' operating income. From 2015 to 2018, the adjustment of the real economy under the new normal of the economy had reduced the employment rate of rural migrant labor and reduced wage income. Since the farmers' operating income still mainly depends on agriculture for a long time, with the rapid decline in the number of rural employment and the continuous decrease of the proportion of agricultural added value in GDP, resulting in slow growth in farmers' operating income. From 2004 to 2010, the average annual growth rate of farmers' property income and the average annual growth rate of transfer income rose steadily; from 2011 to 2018, there was a downward trend, but both were above 11%.

Model Setting
In this paper, 31 provincial-level panel data are selected, pass Hausman's test, then use a two-way fixed-effects model to perform regression. The specific form is as follows: In the Equation, income it represents the income of farmers in the i-th prov- ε is the random error term; C, α, β, and σ are the parameter to be estimated.

Explained Variables
The total income of farmers and its components (income). Select total farmers' income (income 0), farmers' wage income (income 1), farmers' operating income (income 2), farmers' property income (income 3), and farmers' transfer income (income 4) as variables reflecting the income status of farmers.

Core Explanatory Variables
Citilization rate (city) and townization rate (town). The ratio of city population to the total population is used to express the citilization rate, and the ratio of town population to the total population is used to express the townization rate.

Control Variables
According to the existing research results, this article selects corresponding indicators as control variables from three aspects: policy environment, human capital and factor input. In terms of the policy environment, the indicator of financial support for agriculture (fin) is selected, which is expressed by the intensity of the provincial financial support for agriculture, that is, the ratio of agricultural, forestry, and water expenditure to general financial expenditure. In terms of human capital, the educational level (edu) of rural residents and the labor transfer level (trans) are selected. The educational level of rural residents is expressed by the ratio of the number of people with a high school education and above to the total rural population over 6 years old; the labor transfer level is expressed by the ratio of employees in the primary industry to the total number of employees. In terms of factor input, the level of agricultural mechanization (pmac) and the per capita arable land area (pcla) are selected. The level of agricultural mechanization is expressed by the total power of agricultural machinery per capita; the per capita arable land area is expressed by the ratio of the total arable land area to the total rural population.

The Regression Results of the Overall Level of Farmers' Income
This paper uses a two-way fixed effect model to conduct empirical regression to examine the impact of citilization rate and townization rate on the farmers' total income (income 0). The regression results are shown in the following table (Table 4). Model (1) is the result of not adding control variables, and models (2)-(6) are the results of adding control variables one by one.  It can be seen from the above table that when the control variables are not added, the regression coefficient of the citilization rate to the farmers' total income is 0.975. Through the 5% significance test, it shows that every 1% increase in the citilization rate, the farmers' total income will increase by 0.975%. The regression coefficient of the townization rate to the farmers' total income is 1.253, and through the 1% significance test. It shows that every 1% increase in the townization rate, the farmers' total income will increase by 1.253%. "Townization" is more effective than "citilization". The reason is that in recent years, the state has continuously improved relevant policies for the construction of small towns, accelerated the establishment of a high-quality development mechanism for characteristic towns and characteristic small towns, and promoted the development of local urbanization, so that farmers do not have to move to big cities blindly.
Relying on the economic benefits brought by the development of local urbanization, it can also promote farmers' income. After adding the control variables one by one, the regression coefficient of the citilization rate changed from 0.975 to 0.737, and the regression coefficient of the townization rate changed from 1.253 to 0.980. The significance of the regression results of the two remains unchanged, the coefficients changes little, and the empirical results are more steady. From the perspective of control variables, the degree of labor transfer and the level of agricultural mechanization have a positive effect on the total income of farmers, and both are conducive to increasing farmers' income. However, the regression coefficient of the level of agricultural mechanization is relatively small. The possible reason is that certain regions are restricted by geographical environment and economic conditions, which cannot carry out large-scale mechanized production. In most areas, agricultural production adopts cross regional operation, agricultural machinery cooperatives, and other agricultural mechanization modes, which leads to the low level of per capita mechanization and has little impact on the total income of farmers.

The Regression Results of Farmers' Income Structure
In order to examine the impact of the urbanization structure on the income structure of farmers in more detail, this paper examines the relationship between "citilization" and "townization" and the different income components of farmers. The estimated results are shown in Table 5.
It can be seen from Table 5 that "citilization" and "townization" have a and townization rate on farmers' property income are 5.393 and 6.546 respectively, and the regression coefficients on farmers' transfer income are 6.700 and 6.862 respectively, and both pass the 1% significance test. It shows that the development of "citilization" and "townization" can promote the increase of farmers' property income and transfer income, and the promotion of "townization" is greater than that of "citilization". With the development of urbanization, farmers can not only obtain transfer income from various national policies of supporting agriculture and benefiting agriculture, but the rapid development of rural public welfare undertakings has reduced farmers' consumption expenditure on public products and indirectly increased their transfer income. At the same time, the development of urbanization has allowed the circulation of rural land.
Farmers can transfer land contractual management rights through various methods such as sublease and share contracts, and then obtain property lease income. In addition, property income and transfer income only account for a relatively small share of farmers' income, and a series of impacts brought by the development of urbanization will be more obvious in these two types of income.
From the regression results of wage income and operating income, the regression coefficients of citilization rate and townization rate to wage income are −0.379 and 0.125, respectively, and the regression coefficients of operating income are −0.005 and −0.021, respectively. And the test results are not significant, it shows the effects of "citilization" and "townization" on the two are not obvious. In terms of wage income, with the continuous progress and application of modern science and technology, the difficulty of non-agricultural employment for migrant workers with low education levels has continued to increase, and the growth of wage income has slowed down. The inflow of high-quality rural labor to cities and towns has reduced the input of labor factors in agricultural production. In addition, the long-term and extensive agricultural business model has been squeezed by the continuous downward pressure of the "price ceiling" and the rising "cost floor", as well as the continuous approach of the "yellow line" of agricultural production price subsidies and the continuous lighting of the "red light" of the resource environment. The dual constraints have slowed the growth of farmer households' operating income and reduced the room for growth. In addition, the citilization rate and townization rate obtained in this article according to the usual definitions are only the "citilization" and "townization" of the population, while population urbanization is only a necessary condition for increasing farmers' wage income and operating income, not a necessary and sufficient condition.

Conclusions and Policy Recommendations
Based on the provincial panel data from 2007 to 2018, this paper empirically China on farmers' income, and draws the following main conclusions: first, both "citilization" and "townization" can significantly promote the growth of farmers' income, no matter from the perspective of farmers' total income or the components of total income, "townization" has a greater impact than "citilization".
Second, "citilization" and "townization" have an obvious positive effect on farmers' property income and transfer income, especially on transfer income.
Third, "citilization" and "townization" have a negative effect on farmers' wage income and operating income, but this effect is not significant.
According to the above research conclusions, the following policy recom- Due to the complex mechanism of the impact of urbanization on farmers' income and the various paths, the long-term effect of urbanization on farmers' income needs to be further studied. Due to the different economic development conditions and geographical locations of various regions, whether there are differences in the effects of regional urbanization development on farmers' incomes still needs to be discussed in depth. In addition, how to use the dividends brought by urbanization to improve the living standards of farmers, improve the current situation of excessive income gap between urban and rural residents, and promote high-quality economic development, and many other issues require further in-depth research.

Fund Project
The 2019 Graduate Research and Innovation Fund Project "Research on the Impact of the Process of 'Citilization' and 'Townization' on the Increase of Farmers' Income" (ACYC2019160).