The theory of rural industrial convergence can be traced back to the “six industries “theory advocated by Japanese scholar Imamura in the 1990s. He believes that to enhance the added value of agricultural products and increase farmers’ income, it is necessary to rely on agriculture to do a good job in the integrated development of agricultural “production, processing and marketing” to resolve the dilemma of Japanese agricultural development (Li et al., 2020). The central idea of the “six industrialisation concept” highlights the essential role of agriculture and reflects the importance of rural industrial convergence for regional economic benefits (Qi et al., 2021). Rural industrial convergence is essential for China to implement a rural revitalisation strategy and achieve shared prosperity (Fu et al., 2022). In January 2016, the General Office of the State Council officially issued the document “Guiding Opinions on Promoting the Convergence and Development of Rural Primary, Secondary, and Tertiary Industries” (Central People’s Government of the People’s Republic of China Network, 2016). At this point, the convergence of rural industries has become a critical development way to sustain farmers’ income in China. Currently, the literature on rural industrial integration and farmers ‘income growth focuses on the following two aspects:

First, the impact of rural industrial integration on farmers’ income growth. One hand, rural industrial convergence has a significant positive impact on farmers’ income (Wang and Li, 2019). The convergence of rural industries helps alleviate farmers’ poverty vulnerability (Li and Lu, 2019). It promotes farmers’ income growth by extending the agricultural industry chain, giving full play to agricultural versatility, and promoting the integrated development of agricultural service industries (Tang and Hu, 2017). The income of the subjects participating in the convergence of rural industries has increased significantly (Guo et al., 2019; Li et al., 2020). In particular, the average gross and operating incomes of farm households involved in rural industrial convergence are significantly higher than those of households not involved in rural industrial convergence (Yang and Ding, 2019). The micro-survey data of farmers’ households also confirmed that compared with the traditional agricultural development model, the income increase effect of farmers participating in rural industrial convergence is as high as 50% (Li et al., 2017).

On the other hand, there is heterogeneity in the impact of rural industrial convergence on farmers. The impact of rural industrial convergence on farmers is not the same, and the effect of low-income farmers’ income increase is more prominent (Qi et al., 2021). In addition, due to the differences in location conditions, resource conditions, development foundations and other factors between rural areas (Li J. J., 2021), the impact of rural industrial integration on farmers’ income between different regions is also heterogeneous (Wang and Li, 2019).

Second, rural industrial convergence helps to narrow the income gap between urban and rural areas and can lead farmers to shared prosperity. Agricultural industrialisation is the focus of rural industrial convergence. It is a crucial breakthrough to narrow the urban–rural income gap (Chen, 2005). The critical reason for the increasing urban–rural income gap in Hunan Province is that agricultural industrialisation is not high (Song, 2011). The impact of urban–rural income gap changes in the Yangtze River Economic Belt is also in developing rural industrial convergence (Li, 2022). Promoting agricultural industrialisation, improving land and labour productivity, and increasing farmers’ operating income will help narrow the urban–rural income gap (Lai, 2012).

When it comes to industrial convergence, it is necessary to analyse urbanisation construction because industrial development and urbanisation are interactive, and industrial employment structure can significantly affect the urbanisation rate (Long et al., 2015). In the early days, Lewis (1954) found that the industrial sector absorbed surplus labour from the agricultural sector and played an essential role in increasing the income of rural labour. Rural workers are usually affected by the expected income gap to decide whether to migrate from rural to urban areas (Todaro, 1969). Urbanisation is gathering population and resources into cities and towns, promoting social and cultural convergence and economic structure optimisation (Liu et al., 2023). With the continuous improvement of the economy and industrial structure, farmers’ income has been increasing. The literature research on rural industrial convergence and urbanisation development mainly focuses on the following:

First, rural industrial convergence and urbanisation development have a promoting effect. Michaels et al. (2008) believe that with the advancement of urbanisation, the specialisation of the labour force improves production efficiency and promotes the improvement of technological innovation level and the aggregation of emerging industries, which leads to the adjustment of industrial structure. Duan (2017) found that the key to the success of urbanisation is to realise the upgrading of industrial structure through the improvement of production efficiency; the adjustment of industrial structure, especially the change of the proportion of employment structure, is the reason for the increase of urbanisation rate (Sun and Chai, 2012; Hong, 2013). Similarly, urbanisation can promote the transformation of industrial and employment structures (Li, 2011).

Second, there is a negative correlation between rural industrial convergence and urbanisation development. Farhana et al. (2014) found that due to the division of labour under urbanisation, developing countries promote urbanisation in an extensive development mode, which inhibits the adjustment and optimisation of industrial structures. When the level of new urbanization does not match the relative amount of industrial transfer, such as industrial transfer lags behind the development of new urbanization, and the development of new urbanization inhibits the transfer of industrial structure (Zhou et al., 2019). With regard to urbanization, different countries and regions have different impacts on their industrial structure in developing countries. There is no significant impact on its industrial structure in Africa (Gollin et al., 2015). In addition, urbanization will also have a negative impact on some industries. For example, urbanization will have a specific crowding-out effect on manufacturing (Huang and Qiu, 2017). At the same time, other factors (such as the level of financial support for agriculture) will also restrict the effect of urbanisation on the development of rural industrial convergence (Li X. L., 2021).

Based on the current literature review, it is found that most researchers focus on the analysis of the single factor influence of rural industrial convergence or urbanisation development level on farmers’ income or directly discuss the developing relationship between rural industrial convergence and urbanisation. Few people include rural industrial convergence, urbanisation development level, and farmers’ income growth in a research framework. Therefore, it is one of the marginal contributions of this paper to integrate the above three into an analytical framework and use the level of urbanisation development as an intermediary variable and threshold variable to carry out empirical research on the intermediary effect and threshold effect. The second marginal contribution is constructing the index system of rural industrial convergence and using the principal component analysis method to objectively measure rural industrial convergence development.

The influence mechanism of rural industrial convergence on farmers’ income is as follows: First, rural industrial convergence is conducive to developing agricultural industrialisation and improving comprehensive agricultural income. Jiang (2017) believes that agricultural industrialisation is the source and main content of the convergence and development of rural industries. Agricultural industrialisation focuses on the radiation and driving role of leading enterprises and realises the endogenous development of rural areas and the multi-functional development of agriculture (Zhao, 2015). In particular, the convergence of primary, secondary, and tertiary industries led by farmers’ professional cooperatives can enhance the potential of agriculture itself, realise the increase of agricultural added value and sustainable development, and increase farmers’ income (Li and Liu, 2019). Second, the convergence of rural industries is conducive to broadening the channels for farmers to increase their income and build a non-agricultural employment platform. (Rhodes, 1993) believes that industrial convergence makes the production and sales stages of agricultural products and other related products more integrated, forming an orderly chain from the supply of production materials to the processing and retail of products so that farmers can obtain the profits of industrialisation. China’s rural industrial convergence development practice models mainly include leading agricultural enterprises, industrial and commercial capital, vertical convergence management, and ‘Internet + agriculture’ e-commerce platforms (Lv and Liu, 2017). These platforms save production transaction costs and create conditions for realising local non-agricultural employment and household utility maximisation (Cai et al., 2020). The income of farmers has increased steadily in the development trend of rural industrial convergence. Because of this, this paper proposes hypothesis 1.

The level of urbanisation development is an indirect mechanism for rural industrial convergence to promote farmers’ income increase. It is proposed to optimise the industrial structure based on urbanisation (Tian and Abdurezak, 2010), and the convergence of rural industries can also promote the development of local urbanisation (Li and Zhao, 2017). In the long run, there is a positive interaction between urbanisation development and farmers’ income growth (Song and Xiao, 2005). The impact of urbanisation development on farmers’ income growth: First, urbanisation development is conducive to improving the operating income of rural households. Population urbanisation is an essential manifestation of urbanisation development. With the migration of agricultural surplus labour, the requirements of agricultural production for agricultural production are also increased, thus increasing the operating income of farmers (Li, 2016). At the same time, by transferring farmers’ land use rights, we can expand the scale of farmers, thus creating new development opportunities for farmers to increase their income. Second, the development of urbanisation is conducive to improving the wage income of rural residents. Urbanisation helps to increase farmers’ wage income (Wang and Peng, 2013). Urbanisation development provides farmers with many employment opportunities, and the transfer of rural labour force employment increases wage income (Fang and Zhang, 2015). Thirdly, urbanisation development is conducive to improving farmers’ financial transfer payment income. Urbanisation development is closely related to local economic growth. ‘As urbanisation enters a stage of rapid development, a large number of rural surplus labour force pours into cities and towns, the level of industrialisation continues to increase, and economic growth accelerates simultaneously (Cai et al., 2023). When a region’s economy develops to a certain extent, its funds for supporting agriculture will be adequately guaranteed. Simultaneously, there will be more capital inflows of transfer payments to rural areas. Based on this, this paper proposes Hypothesis 2.

The proportion of the urban population is significantly related to the industrial structure and distribution (Long et al., 2015) and plays an essential role in the city’s economic development. Of course, there is a non-linear relationship between industrial integration and regional economic development. For example, the convergence of cultural and tourism industries can positively promote economic growth in ethnic areas, and its impact has non-linear characteristics. There is an optimal level of convergence of cultural and tourism industries that promotes economic growth (Zhao and Wang, 2022). In addition, urbanisation development also has a threshold effect on farmers’ income growth. Since the mid-1980s, farmers’ income has been developing in an unstable and discontinuous trend, and the growth has been prolonged (Li et al., 2017). In recent years, due to the fluctuation of economic development and the weak occupational stability of migrant workers, the difficulty of increasing farmers’ income and the risk of partial reduction have increased significantly (Jiang and Lu, 2017). Therefore, this paper proposes hypothesis 3.

On this basis, this paper examines the relationship between the degree of urbanisation on rural industrial agglomeration and the increase in farmers’ income, examines the regulatory effect of the degree of urbanisation on rural industrial agglomeration and peasant household income, and takes urbanisation as an intermediary variable. This part mainly refers to the literature of Baron and Kenny (1986) and Wen et al. (2004), then introduces the stepwise test method to construct the intermediary effect model.

In equation (1), Income_i,t refers to the disposable income of farmers in i province during the t period, Industry_i,t refers to the development level of rural industrial integration in i province during the t period, Z_i,t is a group of control variables,ℇ_i,t is a random disturbance term. Based on the test of the influence coefficient of rural industrial convergence development level Industry_i,t on farmers’ income increase Income_i,t, β₁ in Eq. 1. The linear regression equation of rural industrial convergence development level Industry_i,t to the intermediary variable urbanisation development level Urban_i,t is constructed (such as Eq. 2). Finally, the linear regression equation of Industry_i,t and intermediary variable Urban_i,t to Income_i,t (such as formula 3) was constructed, and the significance of regression coefficients such as α₁, λ₁, and λ₂ judged the existence of the intermediary effect. In addition, to further explore the non-linear effect of agricultural industry convergence on farmers’ income due to urbanisation development, a threshold effect model of agricultural industry convergence on farmers’ income under the background of urbanisation construction is constructed to verify Hypothesis 3. This paper draws on Hansen (1999), Wang (2015) and other threshold research experience, combined with the actual data, to set up a double threshold model; the particular assumptions are as follows:

i represents the region, t represents the year, Income_i,t is the explained variable, μ_i is used to reflect the individual effect of the region, Urban_i,t represents the threshold variable, γ as the specific threshold value, I(.) is the indicative function, and the true and false are determined according to the results of 0 and 1 in the brackets. Z_i,t is a set of control variables, θ is the corresponding coefficient vector and ℇ_i,t is the random disturbance term.

There are 493 observed variables in this paper, and the descriptive statistics of variables are shown in Table 3. Among them, the mean value of the explained variable farmers’ income is 42.53329, the standard deviation is 16.35281, the minimum value is 17.96, and the maximum is 98.68276. These data show that the income gap between provinces is narrowing with the steady growth of farmers’ per capita disposable income. The mean value of the comprehensive development level of rural industrial integration is 1.1, the standard deviation is 0.5671242, the minimum value is 0.0965812, and the maximum value is 3.784256. This means that the comprehensive development level of China’s rural industrial integration has a large gap between provinces, with a difference of 39 times between the maximum and the minimum. The development of rural industrial integration still has a long way to go. In addition, the development of urbanization is relatively balanced. Still, the maximum and minimum values of variables such as government financial support, scientific and technological innovation, domestic tourism development level, and opening up are far from each other.

This paper verifies the multivariate collinearity among the explanatory variables by using the expansion factor. The judgment method is based on the V.I.F. of the explanatory variable. When the V.I.F. exceeds 10, the multiple correlations are higher. The maximum value of V.I.F. is 6.16 from the detected value of the variation expansion factor in Table 4. It can be seen that there is no significant multivariate co-linearity among the explanatory variables. In addition, to prove that each variable is related to the other, this paper has carried out the correlation coefficient test, which also has a specific correlation.

This paper estimates the panel data by ordinary least squares linear estimation (O.L.S.). Firstly, it empirically studies the relationship between industrial convergence and farmers’ income. On this basis, the control variables are added sequentially, and finally, the mode of (1) ~ (5) is obtained. As shown in Table 5, the regression analysis of the model (5) shows that for rural industrial integration, each additional 1 percentage point will bring 17.32 percentage points of growth to farmers. Therefore, rural residents’ per capita disposable income will increase with the improvement of rural industrial integration. This result confirms that the “rural industry integration” mentioned in Hypothesis 1 plays a positive role in promoting the economic development of farmers. Among the control variables, the degree of government support, openness, and the development of domestic tourism is most closely related to the economic income of farmers. Among them, the opening-up index is positive, indicating that the degree of openness of enterprises is higher, and the degree of internationalisation of their products is higher; the development of enterprises has improved, and farmers have obtained more jobs; their treatment has been better guaranteed, and their quality of life has been further improved. In addition, the coefficient of domestic per capita tourism consumption is positive, which shows that under the influence of the improvement of domestic per capita tourism consumption and the policy of ‘tourism +’ industrial integration, farmers’ income will also rise gradually.

This paper draws on Baron and Kenny (1986) and Wen et al. (2004) three-step method to analyse the intermediary effect empirically and find an indirect transmission mechanism between rural industrial convergence and farmers’ income. Urbanisation is the intermediary variable between the two, and the result is significant. The mediating effect test is divided into three steps: the first step, through the β₁ coefficient in equation 1 to represent the total effect of rural industrial convergence on farmers’ income, is clear that the β₁ coefficient is significantly positive. In the second stage, an empirical analysis was conducted to analyse the impact of rural industrial integration on urbanisation development according to Eq. 2. Model (6) shows that the variable Urban is significantly positive at 1%, indicating that rural industrial convergence is rising and urbanisation development is gradually accelerating. The third step is to test whether the results of rural industrial convergence (Industry) variables in Formula 3 are significant. Model (7) shows that the empirical results of rural industrial convergence (Industry) are significant, and the correlation coefficient decreases, which indicates that there is a partial mediating effect between urbanisation development (Urban) and rural industrial convergence (Industry). The Sobel test was carried out simultaneously for the validity of the above results. It was found that the Sobel test p value was 0.0004, the Z value was 3.485, and the p value of the test results was less than 0.001. In addition, the two significance tests of Goodman1 and Goodman2 also meet the requirements, which are significant at 1%. Alternatively, the proportion of mediating effect to total effect is 28.52%, and the empirical conclusion supports Hypothesis 2 (see Table 6).

Using the development degree of urbanisation as the threshold variable and the panel threshold regression method, this paper examines the internal relationship between rural industrial convergence and the increase in farmers’ income. It reveals the mechanism of rural industrial convergence on increasing peasant household income. To estimate all the parameters of the threshold model, the research should first test the number of thresholds and credibility. Firstly, the number of threshold values of the model is searched and determined, and the F value and p value of a single threshold and two thresholds are obtained, as shown in Table 7. It can be seen from Table 7 that the single threshold is not significant, the double threshold effect is significant at the level of 10%, and the corresponding p values are 0.0940. Therefore, this paper selects the double threshold model for analysis.

To detect the authenticity of the threshold value, we construct a confidence interval to judge. This paper uses a double threshold estimate, and the result is inferred by a 95% confidence interval (as shown in Table 8). The paper also uses likelihood ratio function graphs, such as Figures 1, 2. Since there are two threshold values, it is a segmented display, which can help us understand the threshold value’s estimation and the confidence interval’s construction process. The estimated value of the threshold parameter refers to the value of γ when the likelihood ratio test statistic L.R. = 0, which is 0.7960 (Figure 1) and 0.8500 (Figure 2) in the double threshold model of this study.

From the threshold effect regression result in Table 9, it can be seen that the impact of rural industrial convergence on farmers’ income changes in three intervals. Rural industrial convergence positively impacts farmers’ income growth by acting on urbanisation, which verifies hypothesis 3. According to the double threshold test value of urbanisation, rural industrial convergence’s impact on farmers’ income growth is significantly positive when urbanisation is less than or equal to 0.7960 (Urban ≤0.7960). For every point of growth in rural industrial convergence, farmers’ income will rise by 6.311 points; when the level of urbanisation is between 0.7960 and 0.8500 (0.7960 < Urban ≤ 0.8500), the income of farmers will rise by 8.831 points for every increase in rural industrial convergence. Indeed, as the fundamental driving force to promote the development of urbanisation, the Industry creates opportunities for farmers’ employment through industrial agglomeration, industrial chain extension, and industrial multi-functional expansion (Zhu and Zhang, 2022). The flow of farmers accelerates urbanisation construction, and at the same time, the disposable income of farmers increases. In addition, from Model (8), it can be seen that the influence coefficient of rural industrial convergence on farmers’ income increases first and then decreases. When the urbanisation level is between 0.7960 and 0.8500, the coefficient of rural industrial convergence affecting farmers’ income growth is the highest.

Based on the national and regional development strategy and the geographical location of each province, drawing on Shen et al. (2021), three groups of regional division criteria for eastern, central, and western¹. This paper presents a comparative study of 29 provinces in eastern, central, and western China. As can be seen from models (9) to (11), the results of the impact of the integration of the “three industries” in rural areas on farmers’ income growth are significant at the 1% level in both the eastern and western regions. At the same time, they are significant at the 10% level in the central region. Among the three regions, the eastern region has the best result for the impact of the integration of rural “three industries” on farmers’ income, which is related to its good agricultural industry base; the western region ranks second, which mainly depends on the strong support from the Chinese government to the western region; the coefficient of the central region is relatively low, which indicates that there is still room for development in the impact of rural industrial convergence on farmers’ growth. Besides, the coefficient ranking of urbanisation development affecting farmers’ income growth is consistent with the coefficient ranking of the impact of rural industrial convergence. For the control variable, its impact on farmers’ income is not consistent in different models (see Table 10).

In the empirical robustness study, we use a model about the fixed effect of the relationship between industrial convergence and farmers’ income (as shown in Table 11), as follows: First, it adjusts the sample size and deletes the municipalities to carry out the robustness test. Since Shanghai, Beijing, Chongqing, and Tianjin are municipalities directly under the central government, there are differences between urban volume and provinces. Thus, this paper directly excludes these four municipalities for regression. Model (12) shows that the impact of rural industrial convergence on farmers’ income is still significant. Second, The study is robust by lagging the explanatory variables one period; the lag of one explanatory variable, L.industry in the model (13), passes the significance test at the 1% level, and the coefficient is positive, close to the benchmark result. Third, robustness is measured by replacing control variables. In model (14), the degree of economic development (Economy), the degree of human capital (Capital), and foreign investment (FDI) are added. Through linear regression analysis of the explanatory variables, the regression coefficients are robust and remain significantly positive at the 1% level. In terms of the control variables, the level of G.D.P. per capita (Economy) has a significant effect on farmers’ income with a positive coefficient; the number of college students per 100 people (Capital) also has a significant effect on farmers’ income with a negative coefficient; The effect of foreign investment, i.e., total investment in foreign enterprises, on farmers’ income is significantly positive. The results of this study are robust through the above three methods.

This study conducted empirical research on developing rural industrial integration in 29 provinces (municipals) in China from 2004 to 2020. Least squares, mediation models, and panel threshold models were used to investigate the intrinsic relationship between the development of rural industrial integration and urbanisation and farmers’ income growth. The research conclusions are: (1) Rural industrial convergence is essential to increase farmers’ income. In addition, the empirical results are still robust by deleting the samples of municipalities, lagging one-period explanatory variables, and replacing control variables. (2) The degree of urbanisation as an intermediary variable of rural industrial convergence development affecting farmers’ income increase indirectly promotes the growth of farmers’ income, and the intermediary effect is significant. (3) The level of rural industrial convergence development has a non-linear impact on farmers’ income. The degree of urbanisation development is the threshold variable and passes the double threshold test. The impact of rural industrial convergence development on farmers’ income increases first and then decreases. When the level of urbanisation is between 0.7960 and 0.8500, rural industrial convergence development has the highest impact on farmers’ income. (4) Rural industrial integration on the increase of farmers’ income has regional differences. The contribution of rural industrial integration to farmers’ income growth was significantly positive at the 1% level in both the eastern and western regions, with the highest coefficient of influence in the eastern region and the second highest in the western region. Provinces in the central region were significantly positive at the 10% level.

Based on the above research conclusions, it can get the following policy implications:

First, it’s necessary to give full play to the advantages of rural industrial convergence to improve the income structure of farmers and provide sustainable development momentum for farmers’ income growth. One is that it should highlight the essential functions of agriculture, develop modern agriculture, and increase farmers’ operating income. Each region should develop a modern industrial system of agricultural products according to local conditions in combination with regional resource endowment differences, not only to ensure the adequate supply of agricultural products but also to avoid the homogenisation of agricultural industry development, form a reasonable layout of the agricultural industry and agricultural characteristic brand, and realise the steady increase of farmers’ operating income. The other is it should strengthen the function expansion of the secondary and tertiary industries of agriculture and broaden the wage income of farmers. Fine agricultural industry division of labour, extending the industrial chain through the rich agricultural production, processing, sales, circulation, and another modern agricultural support system, using rural cooperatives or leading agricultural enterprises and other organisations to radiate the leading role of non-agricultural jobs for farmers. Another is to improve the national financial support for agriculture and agricultural policies and increase farmers’ financial transfer payment income.

Second, it should use the advantages of “city-industry convergence” and take the opportunity of county urbanisation construction to provide new development impetus for farmers’ income increase. According to the National Bureau of Statistics of China’s urbanization rate of 0.6522 in 2022, the impact of China’s urbanization development level on farmers’ income growth has not yet reached the optimal level. At present, China’s urbanization development strategy has shifted from the provincial level to the county level. Therefore, it is essential to strengthen the integrated development of county urbanisation and rural industry, guide the rational flow and effective convergence of urban and rural industrial resource elements, and establish a rural industrial development system with the county as the development centre. One of them is it should create a relaxed industrial development environment in the process of county urbanisation construction, break through the bottleneck of rural industrial convergence development, and create a good development environment for farmers’ income increase. The second one is it should resolve the dilemma of lagging transformation of rural industrial production and processing, resource and environmental constraints, financial and insurance support, and market channel development with new ideas, new methods, and new measures. Then, it needs to take the opportunity of in-situ urbanisation development in the county to provide farmers with more non-agricultural jobs and expand farmers’ income channels.

Third, it is essential to give full play to the advantages of industrial development in the eastern, central, and western regions and formulate diversified and differentiated policies for increasing farmers’ income. At first, It should learn from advanced model experience. The eastern region has a good industrial foundation and preferential regional development policies. The impact coefficient of rural industrial convergence on farmers’ income growth is the highest. So, summarising the typical experience of the impact of rural industrial convergence on farmers’ income growth in the eastern region provides a reference for the central and western regions. The second is to stimulate the development potential of the central region. Most of the agricultural provinces in the central region are innovative in the mechanism of rural industrial convergence in the central region, giving full play to the advantages of resource endowments in the central agricultural provinces, taking the road of convergence of the three industries in the agricultural provinces, and maximising the disposable income of rural residents. Lastly, it should use the advantages of the Western development policy to store the growth momentum of farmers’ income in the Western region.

Fourth, It is essential to rely on information technology to promote the convergence of rural industries and broaden the channels of increasing farmers’ income in any way. The rapid development of science and technology promotes the improvement of agricultural production efficiency and broadens the channels for the growth of farmers’ agricultural and non-agricultural income. Based on this, on the one hand, It should complete the construction of agricultural modernization as soon as possible with the help of information technology and increase farmers’ income through scale, organization and intensification; on the other hand, it should extend the agricultural chain using the communication function of information technology (such as live broadcast, small video, etc.), meanwhile making use of ‘rural tourism +’ policy to expand the function of agricultural, and steadily increase farmers’ income.