The reform of the rural land system has profoundly influenced socio-economic development in China. Throughout the periods of socialist revolution, construction, reform, opening-up, and modernization, it has served as a powerful catalyst for advancing productive forces and adjusting production relations. However, as socialism with Chinese characteristics enters a new era, the differentiated urban–rural land system has limited rural development and widened the urban–rural gap. The limitation arises from restrictions on the direct market trading of rural collective construction land, which hampers the ability to meet current needs of high-quality economic development. Therefore, allowing RCCLM is an effective strategy to create a unified urban–rural construction land market and reduce imbalances in urban–rural development, making it a key focus of China’s rural land system reform (Wang et al., 2023). Based on this, the Third Plenary Session of the 18th CPC Central Committee in 2013 adopted the Decision on Several Major Issues Concerning Comprehensively Deepening the Reform. This decision proposed that, under the premise of meeting planning and use controls, rural construction land be allowed to transfer, lease, or turned into shareholdings, placing it on the same market with equal rights and pricing as state-owned construction land. In 2015, the 13th session of the 12th National People’s Congress Standing Committee approved the Decision on Authorizing the State Council to Temporarily Adjust the Implementation of Relevant Legal Provisions in the Administrative Regions of Beijing Daxing District and 32 Other Pilot Counties (Cities and Districts). It selected 15 regions to undertake pilot reforms for RCCLM. In September 2016, this reform was extended to all 33 pilot areas. The 2019 amendment to the Land Management Law officially granted market participation rights to rural collective construction land. However, nationwide implementation has been limited. Both the 27th meeting of the Central Comprehensive Deepening Reform Commission in September 2022 and the subsequent Opinions on Deepening the Pilot Program for the Market Entry of Rural Collective Operational Construction Land issued by the General Office of the Central Committee of the Communist Party of China and the General Office of the State Council emphasized a cautious approach to market reform. Consequently, the Ministry of Natural Resources initiated a new round of pilot programs in selected counties nationwide to further explore this reform. The delay in implementation indicates persistent practical challenges and unmet expectations. To fully unlock the potential of institutional reforms, it is imperative to broaden our understanding of the effects of RCCLM. While focusing on how the policy can effectively alleviate the shortage of urban construction land, it is equally important to investigate whether it has an impact on rural industrial integration development.

Rural industrial integration is a complex process that encompasses multiple sectors and dimensions, including agricultural chain extension, multifunctional agriculture, and agri-service integration. RCCLM has revitalized rural land factors, significantly impacting various aspects of this integration. This section examines the relationship between RCCLM and rural industrial integration through direct impacts and transmission mechanisms, proposing relevant research hypotheses.

First, the study treats the RCCLM pilot policy as a quasi-natural experiment, employing a difference-in-differences model to evaluate its impact on rural industrial integration. Pilot counties serve as the treatment group, while non-pilot areas form the control group. Given the policy’s two-phase implementation in 2015 and 2016, a multi-period difference-in-differences model is utilized. The specific model is constructed as follows:

In Equation 1: INTE i , t is the explained variable, indicating the level of rural industrial integration development of i county in the t year; DID i , t is the core explanatory variable, indicating i county in the t year whether belongs to RCCLM pilot area; X i , t is the control variable; μ i , δ t , and ε i , t are county fixed effect, time fixed effect, and random error items, respectively.

Second, in order to investigate whether RCCLM enhances rural industrial integration by impacting technological innovation and entrepreneurial activity, this study employs a mediation effect model. The specific model is constructed as follows:

In the above equations, M E D i , t denotes the mediating variables, including the level of technological innovation and entrepreneurial activity, and the meanings of other variables are consistent with the benchmark model.

Based on the research content of this paper and the above measurement model, the selected variables and the specific measurement methods are described as follows.

Table 3 presents the benchmark regression results of impact of RCCLM on rural industrial integration. Columns (1) through (3) progressively incorporate control variables and account for county and time fixed effects. The core explanatory variable, RCCLM, shows significantly positive coefficients at the 1% level, confirming Hypothesis 1: RCCLM promotes rural industry integration. These results suggest that RCCLM has enhanced rural land value and attracted industrial and commercial capital to rural areas. This development facilitates the extension of agricultural industry chains, leverages multifunctional potential of agriculture, and provides a crucial impetus for strengthening connections between agriculture and secondary/tertiary industries, ultimately fostering their deep integration.

To further investigate impact of RCCLM on rural industrial integration, this study performs separate regressions on three dimensions: agricultural industry chain extension, multifunctional agriculture, and agriculture–service integration. Table 4 shows that RCCLM significantly and positively affects multifunctional agriculture and agriculture–service integration, but not agricultural industry chain extension. This likely stems from the fact that while rural construction land strengthens linkages in agricultural production, processing, and marketing, persistent challenges in rural areas, such as weak industrial foundations and underdeveloped financial services, hinder the influx and clustering of external industrial and commercial enterprises. Consequently, at this stage, RCCLM has not significantly influenced the extension of the agricultural industry chain.

For the estimation results of the multiple-period difference-in-differences model to be valid, the treatment and control groups must satisfy the parallel trend assumption, meaning there should be no significant difference in the level of rural industry integration between the two groups before implementing RCCLM. To test this, the following model is established for the parallel trend test:

In Equation 5, σ s is the coefficient to be estimated, representing the gap between the level of rural industrial integration development in the treatment and control groups before and after the implementation of the policy, s is the number of periods compared to the time of implementation of the policy of RCCLM, and σ 0 is a constant term, and the meanings of other variables are consistent with Equation 1.

The test results are shown in Figure 1. It can be seen that for s < 0, all σ s are not significant. The results show no significant difference in the level of rural industrial integration development between pilot and non-pilot counties before the implementation of RCCLM, indicating that the parallel trend hypothesis is satisfied.

To further eliminate bias from unobservable factors, a placebo test was conducted based on existing literature. An equal number of counties as the pilot group were randomly selected to form a new treatment group, and the double-difference estimation was re-run to obtain erroneous coefficient estimates of the interaction terms. This process was repeated 1,000 times, and the distribution of the 1,000 estimated coefficients and their corresponding p-values were plotted. Figure 2 shows the results of the placebo test, where the distribution of estimated coefficients from random sampling centers around 0, approximating a normal distribution. The estimated coefficient in the benchmark regression is 0.0147, significantly different from the placebo test results, demonstrating the robustness of the benchmark regression results.

PSM-DID. The selection of pilot areas for RCCLM is not completely random. To better utilize the pilot policy, the government may prefer areas with a larger stock of collectively operated construction land or higher economic development, leading to sample selectivity bias. To address this, we use cross-sectional matching and period-by-period matching methods to match propensity scores and test the robustness of the regression results. The specific steps are as follows: (1) set the control variables in this paper as covariates. (2) Perform kernel matching based on cross-sectional and year-by-year matching and exclude non-co-supporting samples from the total sample to obtain two datasets. (3) Rerun the regression using the multi-period DID model. The results, shown in Table 5, indicate that the regression coefficients of RCCLM on the integrated development of rural industries are significantly positive at the 1% level under both matching methods. This further verifies the robustness of the benchmark regression results.

First, accounting for the dynamic factors at the prefecture level, the study incorporates an interaction term between prefecture-level city and year into the baseline model, following Yuan and Xia (2022). The results in column (1) of Table 6 demonstrate that the coefficient of RCCLM remains significantly positive at the 1% level, affirming the robustness of the initial findings against prefecture-level environmental changes.

Second, excluding the impact of other competition policies. From 2010 to 2022, various national policies like the “Broadband China” initiative and rural entrepreneurship programs have been launched, impacting rural industrial integration. The “Broadband China” policy has facilitated digital technology in rural regions, removing technical obstacles for e-commerce, thereby expanding agricultural product sales online, and fostering growth in logistics, packaging, and warehousing. This extends the agricultural industry chain, enhances its value, and thus influences the overall integrated development of rural industries. Furthermore, the policy encouraging rural workers to return home for entrepreneurship offers technology, capital, and tax benefits, boosting local innovation and business start-ups. This initiative supports the emergence of new agricultural formats and entities, impacting rural industrial integration. Upon including these policy effects in the regression model, as shown in column (2) of Table 6, the positive impact of RCCLM on rural development persists, confirming the robustness of baseline findings against other policy influences.

Third, 1% tailing. To mitigate the influence of extreme values on the benchmark regression results, all continuous variables were subjected to a 1% tailing. The re-run of the regression, presented in column (3) of Table 6, shows that the RCCLM regression coefficients remain significant at the 1 percent level, consistent with the benchmark estimates. This indicates that the benchmark results are robust against extreme values.

Fourth, shorten the sample duration. The article initially covers the sample study interval from 2010 to 2022, but since the RCCLM policy identified pilot lists in 2015 and 2016, the authors narrow the sample period to 2012–2020 to minimize the potential impact of errors from years further from the policy shock. As shown in column (4) of Table 6, the regression coefficient of RCCLM on the integrated development of rural industries is 0.0127 and is significant at the 1% level. This indicates that the effect of RCCLM on the integrated development of rural industries is not influenced by the length of the sample period, confirming the robustness of the baseline regression results.

Building on the theoretical analysis from the previous paper, this study introduces technological innovation and entrepreneurial activity as mediating variables to construct a mediating effect model. Following the standard practice of sequential testing for mediating effects, it regresses Equations 2–4 to explore how RCCLM influences the development of rural industrial integration. First, the results in column (1) of Table 7 indicate that RCCLM has a significant promotion effect on the integrated development of rural industries. Second, the impact of RCCLM on technological innovation and entrepreneurial activity is analyzed. As shown in columns (2) and (3) of Table 7, the estimated coefficients for RCCLM are 0.0201 for technological innovation and 0.0766 for entrepreneurial activity, both of which are significant. This indicates that RCCLM promotes both technological innovation and entrepreneurial activity. Finally, the impact of the mediating variables on rural industrial integration is examined. Columns (4) and (5) of Table 7 present regressions that include technological innovation and entrepreneurial activity, respectively, in addition to the variables in column (1). The results reveal that the regression coefficients for both technological innovation and entrepreneurial activity on the integrated development of rural industries are significantly positive at the 1% level. This indicates that RCCLM promotes rural industrial integration development through both technological innovation and entrepreneurial activity. In summary, Hypotheses 2 and 3 are confirmed.

First, considering the unbalanced development across regions in China and the varying resource endowments and economic development models, the impact of RCCLM on rural industrial integration may differ by geographic location. Therefore, the total sample is divided into four regions—East, central, west, and northeast—for separate regression analyses, with the specific estimation results presented in Table 8. RCCLM demonstrates a significant positive impact on rural industrial integration in the eastern, western, and northeastern regions, but not in the central region. The reason for this disparity may be that collective land transfer first emerged in the eastern coastal region, where a relatively mature trading model has developed through long-term exploration and reform. Additionally, the demand for construction land for industrial development is higher in this area. Therefore, within the RCCLM policy framework, rural areas in the east are more likely to achieve industrial agglomeration, driving rural industrial integration. In contrast, the central region faces challenges due to the concentration of collectively operated construction land primarily in the eastern coastal areas. Moreover, to strengthen farmland protection, the government has prohibited the conversion of agricultural land into new construction land. This has made it difficult for most central regions to acquire newly added rural construction land. As a result, the limited scope of RCCLM in the central region has not significantly promoted rural industrial integration. In contrast, local governments in the economically underdeveloped western and northeastern regions may prioritize leveraging RCCLM to overcome existing resource constraints and achieve improvements in livelihoods and economic performance. Additionally, national policies aimed at promoting Western Development and the comprehensive revitalization of the northeast provide favorable conditions for policy implementation in these areas. As a result, RCCLM facilitates the integrated development of rural industries.

Second, this analysis considers the heterogeneity of impact of RCCLM on rural industrial integration based on population size. Counties with a resident population exceeding 1 million are classified as large-scale, while those with a population below 1 million are categorized as medium–small scale. The specific regression results, presented in Table 9, indicate that RCCLM significantly promotes rural industrial integration in large-scale counties, while the effect is not significant in medium–small-scale counties. This disparity may arise because local governments in large-scale areas have a greater demand for industrial development to address employment issues and improve livelihoods. Consequently, when urban construction land supply is insufficient, these governments are more motivated to actively promote RCCLM, allowing more economically efficient enterprises to enter rural areas and facilitating rural industrial integration.

The implementation of RCCLM has accelerated the convergence of industrial and commercial capital, human capital, and technical resources in the pilot areas. While this provides essential support for the integrated development of rural industries, the deepening of RCCLM reform may also generate spillover effects on neighboring areas due to the close geographic and resource connections among villages. Therefore, this paper employs a spatial difference-in-differences model to test this hypothesis. The specific model is constructed as follows:

In Equation 6, ρ represents the spatial autoregressive coefficient, and W is the spatial weight matrix, and in this paper, the neighborhood matrix is used for regression. ∅ 1 and ∅ 2 are the elastic coefficients of the spatial interaction term of the core explanatory variable and the control variable, respectively.

First, the spatial correlation test needs to be carried out by Moran’s I index method. The results are shown in Table 10, under the neighborhood matrix, the Moran’s I indexes of the integrated development of rural industries in the sample counties from 2010 to 2022 are all significantly positive, indicating that the spatial correlation test has been passed.

Second, to select the appropriate measurement model, the article sequentially conducts the LM test, LR test, Hausman test, and Wald test. Based on the test results, the study employs a two-way fixed-effects model with a spatial Durbin specification and difference-in-differences model for estimation, as detailed in Equation 6. Regression analysis is then performed, and the results are presented in Table 11. Under the neighborhood matrix, the spatial autocorrelation regression coefficients, ρ and W × DID, for the integrated development of rural industries are significantly positive. This indicates a notable positive spatial correlation in the integrated development of rural industries across regions and confirms that RCCLM has a significant positive spatial spillover effect on rural industrial integration in neighboring areas.

However, analyzing the spatial spillover effect between regions solely through simple point regression may introduce estimation bias. Therefore, to examine the spatial effect of RCCLM on the integrated development of rural industries more precisely, the coefficients of RCCLM in the SDM model are further decomposed into direct, indirect, and total effects through partial differentiation. The results in Table 11 indicate that the indirect effect of RCCLM on the integrated development of rural industries is significantly positive, suggesting that RCCLM generates strong positive externalities and enhances the level of rural industrial integration in neighboring areas through spatial spillover effects.

Rural industrial integration is a crucial exploration for enhancing the level of rural industrialization and achieving rural revitalization. This paper examines the impact of RCCLM on rural industrial integration and presents the following conclusions: first, pilot areas experienced a significant 1.47% increase in rural industrial integration compared to non-pilot areas. The conclusion remains robust after a series of tests. It highlights the positive economic effects of RCCLM and identifies the key driving factors behind rural industrial integration. Second, county-level technological innovation and entrepreneurial activity serve as mediators in the process through which RCCLM promotes rural industrial integration. This finding clarifies the pathways by which RCCLM drives rural industrial integration. Third, the heterogeneity analysis shows that there is heterogeneity in the impact of RCCLM on rural industrial integration development under different conditions of spatial geographic location and population size. Specifically, it shows significant reform effects in the eastern, western, and northeastern regions as well as in counties with larger population sizes, while it is not significant in the central region and in counties with smaller population sizes. Therefore, when implementing RCCLM, the government should adopt a context-specific approach. Fourth, through the sharing of resource elements, the policy exhibits a positive spatial spillover effect on the development of rural industrial integration in neighboring areas. This further expands the research results of the spatial effect of RCCLM policy.

However, this study has certain limitations. First, collecting data on rural industrial integration at the county level in China are challenging. Professional databases, the China County Statistical Yearbook, and various local statistical yearbooks sometimes do not have completely identical statistic caliber, leading to slight differences in some data. Additionally, to address missing statistical data for certain years in the yearbooks, this paper uses interpolation or averaging methods to fill in the gaps. While these methods reasonably supplement the missing data, they may still result in a decrease in the accuracy of some estimation results. For future research, more in-depth exploration and analysis based on regional heterogeneity may be possible. Specifically, while this paper clarifies the mechanism through which the RCCLM promotes rural industrial integration development, it must be noted that the development situations in different regions and the specific practices of local governments in implementing policies are not completely identical, which means that the underlying mechanisms may also differ. Therefore, future research can explore differentiated paths within different regional contexts. To fully leverage the positive impact of RCCLM on rural industrial integration development, a more comprehensive study and empirical analysis of its basic mechanisms are needed.

First, a sound policy system for RCCLM should be established. On the one hand, a reasonable mechanism for the distribution of land value-added gains should be actively explored, so that, while taking into account the need to increase the motivation of local governments to promote RCCLM, more land value-added gains can be used for agricultural and rural development, increasing the sustainability of rural residents’ earnings and providing financial support for the integrated development of rural industries. On the other hand, local governments should focus on the linkage between policies. In the context of rural homestead system reform, policies should be designed to incentivize farmers to voluntarily withdraw from their compliant homesteads and convert them into rural construction land that can participate in the market. This would further expand the pool of land available for marketization, allowing the policy dividends of RCCLM to be fully leveraged to promote rural industrial integration development.

Second, efforts should be made to promote the development of technological innovation and entrepreneurial activities, and fully leverage their intermediary role in the impact of RCCLM on the integration of rural industries. In terms of technological innovation, local governments should focus on the goal of establishing a unified urban–rural construction land market to enhance the competitiveness of the land market. This would compel enterprises to strengthen their innovation efforts to improve economic performance. Concurrently, local governments should increase investment in the infrastructure surrounding rural collective construction land, and maintain the price advantage of rural construction land in transactions, thereby reducing enterprises’ initial operating costs and incentivizing greater innovation investment. Furthermore, on the basis of identifying the leading industries and existing resource conditions at the county level, innovative technologies should be closely matched with the local rural industrial development plan to provide technical support for the integrated development of rural industries. In terms of enhancing entrepreneurial activities, farmers should be prioritized in the distribution of land value-added income from RCCLM, in order to alleviate the problem of insufficient start-up capital. Additionally, local governments should actively encourage the labor force to return to their hometowns for entrepreneurship. When trading rural construction land, they can set certain conditions to moderately favor high-quality returnees who start businesses, effectively combining the personal capital advantages of these entrepreneurs with the local agricultural development characteristics and resource advantages. This will enhance the success rate of entrepreneurs and inject endogenous impetus into the rural industrial integration development at the county level.

Third, the reform of RCCLM should be promoted according to local conditions. RCCLM has not yet played a significant role in promoting rural industrial integration development in the central region or in small- and medium-sized areas. Therefore, in the central region, in the process of promoting RCCLM, we should avoid “one-size-fits-all,” give local governments a certain degree of discretionary space through moderate decentralization by the central government, allow the adoption of differentiated reforms, moderately expand the source of new construction land, alleviate the predicament of insufficient stock of construction land, and insist on land transactions oriented to the development needs of the regional rural industry, thus promoting rural industrial integration development. For small- and medium-sized areas, local governments should further improve the relevant supporting policies for RCCLM and strengthen the improvement of the land parcels entering the market as well as the construction of the surrounding infrastructure, so as to increase the value of the land parcels entering the market and make them more attractive to enterprises and investor, and thus provide capital and industrial infrastructure support for the integrated development of rural industries.