Introduction

The uneven development of urban and rural areas is the biggest imbalance in many emerging economies, such as China, and is one of the main reasons for the decline of the countryside in the stage of economic growth (Cheng et al., 2020). As modernization accelerated, the benefits of development have not been broadly shared in China, forcing the Chinese authorities to prioritize the solution to this issue by narrowing the gap between urban and rural areas on the agenda (Wang et al., 2019). Since 2004, the Chinese central Government has issued “No. 1 Document” for consecutive years on building a new countryside, abolishing agricultural taxes and exempting tuition and fees, and establishing a new rural cooperative medical system, a rural minimum living security system, and a new rural social endowment insurance system. In recent years, the focus has shifted towards prioritizing agriculture and the countryside, fostering urban-rural integration, and facilitating the seamless flow of production factors between urban and rural regions. How to bridge the divide between urban and rural development, and to achieve the harmonious coexistence and prosperity of “city and township,” has attracted unprecedented attention at present (Jiang et al., 2022).

The concept of protecting the food products’ appellation of origin by using a tag originated from the Lisbon Agreement for the Protection of Appellations of Origin and their International Registration in 1958, giving the first clear definition of agricultural products with the characteristics in their original places, and enabling a tagged protection on their primal eco-geographical features. The geographical indication (GI) had been widely used in European countries, and then spread to the globe. Authorized by the Ministry of Agriculture of the P.R. China and deployed since 1991, the Geographical Indicated Agricultural Product (GIAP)Footnote 1 refers to a kind of eco-label marked on a product shown in Fig. 1, usually an agricultural product or a primal foodstuff, which is characterized by its spatial and geographical origin and distinctive characteristics (Cassago et al., 2021; Liu et al., 2024), has become a typical agricultural policy in ecological product value realization (Dou et al., 2024). The policy of encouragement on GIAP’s industry holds immense practical value in expediting the process of cultivation of ecological agriculture and rural modernization, thereby bridging the divide between urban and rural areas and fostering the high-quality development of the agricultural economy (Xiao et al., 2021). Considering these, the China National Intellectual Property Administration (CNIPA) (CNIPA, 2021) issued a notice on organizing and carrying out the action of GIs to help rural revitalization; soon afterwards, CNIPA made further arrangements for strengthening the guidance of key links in promoting the use of GIs (CNIPA, 2023). So far, local governments have shown increasing interest in the declaration and registration of GIAP, which has emerged as a crucial catalyst for regional economic growth (Qie et al., 2023; Zhao et al., 2014).

Fig. 1
figure 1

A diagrammatic sketch of the Geographical Indicated Agricultural Product (GIAP) mark in China (plotted by the authors by referring to the official specimen).

Full size image

Hubei Province is situated in the core of China, positioned to the north of Dongting Lake and in the middle of the Yangtze River, covering an area of more than 185,900 km2, as illustrated in Fig. 2. With large agricultural output and rich farm products (Du et al., 2023), Hubei is the transportation junction in central China, making it a vital component of development strategies of the “Yangtze River economic zone” and “Triangle of central China” (Chai et al., 2019). In 2021, the GDP of Hubei Province reached 5.3 trillion CNY (1 USD ≈ 7 CNY), ranking 7th in China’s provinces. Due to its favorable geographical conditions with plenty of water resources, Hubei Province is widely famous for its characteristic agricultural products (Liu et al., 2021), drawing a mass of scientific research interest in the field of agriculture and food, including cultivated land usage (Zhu et al., 2022), ecological and social impacts by fertilizer conservation (Liu et al., 2021; Yang et al., 2020), agricultural productivity (Xie et al., 2020; Yang et al., 2023), and low-carbon sustainable farming (Chen et al., 2023; Wang et al., 2021c; Zhang et al., 2021). In addition, the abundant riverine and laky resources enable the researchers to carry out the studies of farm water utilization (Zhou, 2019) and the spatial heterogeneity shaped by the wetlands (Cai et al., 2024).

Fig. 2
figure 2

The location of Hubei, China.

Full size image

There are thresholds for an agricultural product to be verified with GIAP, which requires a procedure to acquire. The procedure of obtaining a GIAP tag includes feature-discovering and initial application by the local government, investigation by authorized agencies in terms of quality and safety, and approval by the National Ministry of Agriculture. Hence, different distinctive products are tagged with GIAP at different times. As displayed by Hubei Intellectual Property Bureau (HIPB)Footnote 2 the first product with the GI of Hubei, China was set up in 2004, the GIAP applying work continues yearly, various types of GIAP in different production areas have been approved for produce and sale in Hubei almost every year colored in Fig. 3, where the areas remain white still do not possess any GIAP tag yet; or summed up in Fig. 4. To date, Hubei has accumulated more than 500 GIAPs and an annual production of over 33 billion CNY. The rapid expansion of the local economy has led to an increase in urbanization in Hubei province, yet the problem of unbalanced development between urban and rural areas has become increasingly apparent as well.

Fig. 3
figure 3

GIAP establishment process in Hubei, China (Marked by authors based on the information provided by HIPB).

Full size image
Fig. 4
figure 4

Number of counties newly implementing the GIAP in Hubei, China.

Full size image

Although the GIAP is a nationwide policy in China, systematic appraisements of this policy are infrequent; this may be due to the far-ranging geographical distribution and various categories of GIAP-tagged products. Focusing on a specific provincial administrative region might be a viable way to preliminarily evaluate its poverty alleviation effects. Unlike the sampling survey in a few areas in other provinces, the Department of Housing and Urban-Rural Development of Hubei Province, in particular, has launched a province-wide survey since 2021, namely “Evaluation and Survey on Rural Construction of Hubei Province,” covering the whole 12 Hubei prefecture-level cities. In each prefecture city of Hubei Province, China, all the villages and towns have been surveyed (except for the main prefecture city zones). During the survey, the questionnaires covering the living standards, agricultural productions (including the sales status of GIAP products), infrastructure situations, and other aspects of rural development were answered by the village cadres and villagers. By combining the survey results and field visits, each survey team will complete the government-commissioned investigation reports, serving as a concernful reference on the government-funded rural construction for the incoming year. All these works reflect the high degree of emphasis on rural development by the Hubei provincial government, making sense of an agricultural policy review for Hubei, China.

Therefore, from the provincial perspective, taking Hubei Province as an example has significant research value, which can be seen from the following aspects: (1) Relative consistency in policy implementation: Within a provincial administrative unit, the application, approval, supervision, and enforcement processes of the GIAP policy are relatively more uniform, reducing the noise of policy effect differences caused by different management approaches of provincial governments. (2) Relative consistency in data quality: Utilizing data from a single province makes it easier to ensure the relative consistency of data sources, statistical definitions, coverage, and quality, reducing the estimation bias that may arise from inter-provincial statistical differences or data gaps in national data, especially in earlier years. (3) Internal Heterogeneity and Representativeness: Hubei Province itself has a high degree of internal heterogeneity, encompassing various terrains such as plains, hills, and mountains, as well as large cities, medium-sized cities, and numerous agricultural counties, along with regions of different economic development levels. This rich internal diversity makes the “control group - treatment group” comparison within the province relatively cleaner, reducing the influence of confounding factors caused by significant inter-provincial differences.

The GIAP fundamentally aims to transform the unique ecological value embedded in rural landscapes and agricultural practices into economic benefits through GI. The implementation of the GIAP policy in local areas is proven to have significantly positive impacts on economic development (Dong et al., 2023; Qie et al., 2023). The institutional framework of GIAP—mandating localized production and often incorporating contractual premium-sharing clauses (e.g., cooperatives, guaranteed price premiums for farmers)—ensures primary producers capture a significant share of the generated brand premium. While urban enterprises may participate in processing and marketing, accruing gains from their specific competencies, the incremental economic rent created by GIAP certification predominantly flows to rural areas. This represents a substantial marginal improvement for farmers transitioning from unbranded commodity markets to premium GIAP markets, explaining the empirical observation of significant rural income growth contrasted with muted urban income effects.

However, facing the current situation of unbalanced development between urban and rural areas, does the implementation of the GIAP policy bring more incremental income to the local area, as well as help narrow the urban-rural gap? Moreover, does regional heterogeneity affect the impact of GIAP policy implementation? Do state assistance policies to poor counties enhance the effect of GIAP? To answer the above questions, this paper employed the Theil index to assess the impact of the GIAP policy on the economic disparity between urban and rural areas. Subsequently, a series of analyses, including multi-time-point difference-in-difference (DID), event study, grouped regression, and interaction terms regression, are conducted for further investigation.

The main objectives of this paper are: (i) Theoretically discussing the mechanism of GIAP policy on rural economy, and accurately evaluating the direct and dynamic effects of GIAP on the uneven development of urban and rural areas; (ii) Effectively alleviating the selectivity bias problem, and carrying out a series of robustness tests to ensure the reliability of the research conclusion; (iii) Examining the differentiated impact of GIAP policy on areas with different geographical locations and economic levels, and deeply analyzing the mechanism of GIAP policy affecting the development of city and countryside integration. In general, this study not only provides empirical evidence for further promoting the implementation of GIAP policy but also gives a dependable measurement method to evaluate the effect of GIAP policy on narrowing the imbalance of urban and rural economic development, while also expanding the research literature on the influence of the urban-rural integrated development model on innovation.

Literature review and research hypotheses

Literature review

Uneven development of urban and rural areas

Existing studies have demonstrated that the distinction between urban and rural areas itself is a basic driving force for social development (Yuan et al., 2018). From this point, it is not necessary and cannot fundamentally eliminate the difference between urban and rural areas in the social and economic structure (Zhang and Lu, 2018). However, the antagonism between urban and rural areas is not conducive to the sustainable development of China’s economy, which is also a factor in the disparity between the urban and rural economies (Sun and Yang, 2022). In that case, it is the general and future trends to promote the transformation of urban and rural areas from “one-way flow” to “bilateral interaction,” and from “urban bias” to “urban-rural integration,” finally to realize urban-rural integrated development (Chen et al., 2021).

Understanding the urban-rural imbalance requires a comprehensive consideration of urban-rural coordination, social history, and culture (Ye et al., 2022b). Despite historical, cultural, and institutional differences, the imbalance between urban and rural development is a difficult problem that all countries have encountered or are currently facing (Kilkenny, 2010). The research focus on urban-rural governance varies greatly, but it has received a lot of attention as a result. For instance, international researchers mainly focus on the sustainable development and environmental issues of urban and rural areas, such as policy (Baffoe et al., 2021), management (da Silva et al., 2017), and well-being (Knickel et al., 2021). While scholars in China prefer to pay attention to urban-rural integration and rural revitalization, especially on urbanization (Ji et al., 2019), land planning (Guo et al., 2019), as well as the movement of population (Liu et al., 2022b). Despite these disparities, collaborative urban-rural governance and promoting sustainable development are topics of common concern and critical consensus both domestically and internationally (Jiang et al., 2022; Tian et al., 2022; Ye et al., 2022a).

Existing studies on GIAP

The protection system of GIAP originated in foreign countries other than China, which has been developed for a long time worldwide. China commenced the GIAP identification policy in this century after entering into the World Trade Organization (Yongshun et al., 2022). Previous research has demonstrated that developing GIAP enhances the competitive advantage of products in the international market, which in turn promotes economic and cultural development (Besah-Adanu et al., 2019; Cei et al., 2021; Török et al., 2020). Additionally, in the production process of GIAPs, it is necessary to effectively organize and manage smaller producers; thus, the local governors and managers have to introduce and perform the scientific and advanced systems for unified management (Yu et al., 2023). Consequently, GIAP producers have gained greater economic advantages, while economic development in local areas in turn contributes to the added value of the GIAPs. (Huysmans and Swinnen, 2019; Teuber, 2011).

Up until now, there have been three main classifications of studies on the role of GIAP policies in promoting regional growth, particularly in rural areas. First, the research on the GIAPs policy demonstrated that GIAPs could effectively enable small production institutions and organizations to carry out unified and centralized production, occupying a considerable market space and thus causing a certain impact on some large-scale enterprises in the economy (Mancini, 2013; Wang et al., 2021a). Secondly, due to the monopoly and reputation advantages of GIAPs, it can exclude non-GIAPs and enter niche markets smoothly, helping producers and local government obtain economic benefits (Bimbo et al., 2023; Wong and Elbegsaikhan, 2020). Thirdly, due to the high geographical dependence of GIAPs, the implementation of GIAP policy could eminently stimulate the overall sales of products, then promoting the employment and living standards in the region (Lou et al., 2023; Prathap, 2020).

Despite this, there is still ample opportunity to advance GIAP to a larger scale. For example, the quality of GI products on the market is uneven, and there is a lack of a standard for standardized management, causing (Ihsaniyati et al., 2022; Kovačević et al., 2022). Moreover, the research, publicity, and investment of GIAPs are inadequate; at the same time, few efforts are put into industrial support, base construction, and brand building of GIAPs (Dou et al., 2022; Parsinezhad and Naeimi, 2022). Overall, it is of great need to assess the economic value of GIAPs to accelerate the implementation and promotion of the GIAP policy.

Policy evaluation method

Policy evaluation plays an important role in improving policy design, overcoming drawbacks and obstacles in policy operation, and fully leveraging policy benefits (Zhou et al., 2022). For a long time, much attention has been attached to understanding whether a policy has achieved its ultimate goal and its impact on one’s own interests, and then many of the methods of policy evaluation are performed for this reason (Zhang et al., 2022). The main policy evaluation methods include: the instrumental variable (IV) method, regression discontinuity design (RDD), synthetic control method (SCM), and DID method (Ye, 2022). Nevertheless, compared to others that have heterogeneity issues (for IV method (Mogstad and Torgovitsky, 2018)), high requirements for geographical uniformity (for RDD method (Choi and Lee, 2017)), and model complexity (for SCM method (Cerulli, 2019)), the DID method offers quantitative results while overcoming the interference of unobserved confounding factors, demonstrating evident benefits (Abadie and Cattaneo, 2018; Meyer, 1995; Wang and Qiu, 2021).

Nowadays, the DID method has been widely used in policy effectiveness evaluation in various fields. For instance, Lin et al. (2023) have conducted an in-depth investigation into the influence of China’s environmental protection tax on governance through the DID approach. Wang et al. (2022a) used the DID model to analyze the promoting effect of internet development on urban green innovation efficiency. Zhang et al. (2022) evaluated the innovation performance of local industrial policies for intelligent connected vehicles using the multi-period (DID) method. In addition, more attention has also been placed on policy evaluations related to housing (Kim and Ko, 2023), air pollution control (Liu et al., 2022a), and industrial transfer (Yu and Zhang, 2022), using the DID methods. Overall, a large number of policy evaluation studies using the DID method have been conducted so far, and this method has been and will continue to be the main support for policy evaluation for researchers in the long term.

Potential contributions of this paper

Although both the GIAP measures and urban-rural inequality have been discussed and verified, their combination research is rare; only Zhang et al. (2023) and Ai et al. (2025) have checked the relationship between GIAP and urban-rural gap in terms of a specific region or a certain level of administrative region. In view of this, this paper focused on a certain region with the county-level panel data, treating GIAP as a Quasi-natural experiment rather than a simple continuous variable, trying to probe into its direct causal relationship through the DID model with heterogeneity analysis and possible policy interaction effect analysis.

Research hypothesis

In China and other developing countries, the core causes of the rural-urban income gap are the low rate of production and the low added value of the farm crops and agricultural products (Gelgo et al., 2023; Patra and Babu, 2023; Yu and Zhao, 2009). To ameliorate this situation, supporting policies on farm crops and agricultural products are requisite and diversified. As mentioned before, the GIAP policy implementation has evidently had a positive effect on local economic growth, especially for rural areas, which may help to narrow the urban-rural development gap to some extent. Moreover, the rural areas benefit from the development of GIAPs’ industries in various aspects, such as stimulating tourism and promoting employment, bringing more incremental income growth in this way. Hence, hypotheses 1.1 and 1.2 (H1.1 and H1.2) are put forward as follows:

H1.1: The GIAP policy implementation can effectively narrow the urban-rural development gap in Hubei, China.

H1.2: The GIAP policy narrows the urban-rural development gap by effectively bringing incremental income growth to rural areas in Hubei, China.

Cei et al. (2020) have found that the GIAP policy in Europe brought out temporal heterogeneity among different farms whose GIAP certification was distinct; the spatial heterogeneity of it may also exist. In fact, the effect of an agricultural policy usually exists in geographical or spatial heterogeneity (Bullock and Salhofer, 2003; Sumner et al., 2010; Yang and Liu, 2021). Likewise, some special conditions may be conducive to the effect of GIAP policy implementation (Xu et al., 2022). For example, the Yangtze River Economic Belt is one of the “three major strategies” implemented by the central government, and its regional economic development has received great attention (Qie et al., 2024). Therefore, whether the geographical location close to the Yangtze River has an impact on the local economic development, and the effect of GIAP policy implementation, to a certain extent. At this point, the core geographical factor should be taken into consideration through H2:

H2: The counties in proximity to the Yangtze River can benefit from greater advantages from the GIAP policy.

Shortening the gap between urban and rural areas has long been concerned by the Chinese government (Liu et al., 2020), especially since 2016, when a plan called “Targeted Poverty Alleviation” was actualized to assist the poverty-stricken counties point-to-point (Yang et al., 2020), which can even go down to households (Yan et al., 2023). With the comprehensive promotion of poverty alleviation and rural revitalization, poverty-stricken counties have received assistance intensively and continuously, which may also help reduce the gap (Zhou et al., 2023) and strengthen the implementation effect of GIAP policies. Hence, hypothesis 3 (H3) can be put forward as follows:

H3: The aid policy on poverty-stricken countries has strengthened the GIAP policy’s effects.

Research design and data description

Empirical model’s principle

The DID model has been broadly adopted in the policy effects analysis, particularly for the pilot policies that have been put into effect in certain regions, as the other non-pilot regions can act as the control group to reflect the policy effects, like a natural science experiment (Abadie and Cattaneo, 2018; Greenstone and Hanna, 2014; Song et al., 2011; Xu et al., 2023; Ye, 2022). By comparing the post-policy differences between the treated groups and the controlled groups after a certain time, the policy effect can then be explicitly reflected both in significance and in the coefficient (Fang et al., 2023; Lee and Kang, 2006; Meyer, 1995; Ye et al., 2020). Normally, a pilot policy is executed at a certain single time point among treated groups, hence a single-time-point DID model can be set as:

$$DID{s}_{it}=Trea{t}_{i}\times Pos{t}_{t}$$
(1)

Where in Eq. (1), i represents the individual i, t represents the time t. Treati refers to the exogenous impact (such as a policy implementation), equals 1 if the individual i has been affected (treatment group) and equals 0 for the unaffected individuals. Postt means the time range for t, if t is at or later than the time point of exogenous, Postt equals 1; if earlier, Postt equals 0. Therefore, the dummy variable DIDsit becomes 1 only for those treated individuals on or after the impact occurred.

For inspecting the degree of the impact, the baseline DID model can be set as:

$${Y}_{it}=\alpha +\beta \cdot DID{s}_{it}+\mathop{\sum }\limits_{j=1}^{n}{\gamma }_{j}Contro{l}_{it}+{\delta }_{i}+{\mu }_{t}+{\varepsilon }_{it}$$
(2)

Where in Eq. (2), Yit is the dependent/explained variable to be examined. α is the constant term of the model. β is the coefficient of the core independent / explaining term DIDsit. Factors other than the exogenous impact that may also affect the dependent variable are identified with control variables, which have been delegated by Controlit. j is the number of each control variable, and γj is the corresponding coefficient. δi and μt denote the fixed effects on individual i and the time t, respectively. εit refers to the error term.

Core research variables

Core explaining variable

Distinctly, the GIAP policy executed in those counties of Hubei, China is identified as the core explaining variable Treati in Eq. (1), we use GIAPi to indicate this policy event in the county i.

In this study, the exogenous shock, GIAP implication, was executed in different years among different counties in Hubei province, which means it is a multi-time-point event rather than a single-time-point one. As Fig. 4 shows, the GIAP policy was inconsistently implemented in 2004, 2006–2012, 2014, and 2015.

Considering this, unlike Postt in Eq. (1), the time range term of GIAP is not only determined by the time t but also by the individual i, necessitating its adjustment into Postit due to its multi-time-point nature.

Hence, referring to the DIDs model with multiple time periods (Callaway and Sant’Anna, 2021; Goodman-Bacon, 2021; Guo and Zhong, 2022; Nunn and Qian, 2011), the core explanatory variable should be adjusted to DIDmit and be expressed as:

$$DID{m}_{it}=GIA{P}_{i}\times Pos{t}_{it}$$
(3)

Core explained variable

Usually, the urban-rural development gap can be reflected by directly comparing the income difference between the urban and rural residents, or by employing an index such as the Gini coefficient or Kakwani Index to quantify it. However, in contrast with an angle of absolute equalitarianism, narrowing the imbalanced urban-rural gap turns out to be more pragmatic sense. The Theil Index (Theil), also known as Theil entropy (Johnston, 1969), proposed by Theil (1967), has become a prevalent approach to weigh the unbalanced development in industrial structure (Conceicao et al., 2001; Shao et al., 2021; Wang et al., 2021b), digital divide (Peng and Dan, 2023; Pick and Nishida, 2015; Vasilescu et al., 2020), resources and eco-environment heterogeneity (Wang et al., 2022b, 2023; Zheng et al., 2022, 2023), Energy inequality (Duro and Padilla, 2011; Ma et al., 2021; Pakrooh et al., 2020; Wang et al., 2021b), unequal regional carbon emissions (Luo et al., 2023; Tian et al., 2021; Zhao et al., 2023), and uneven income gap between urban and rural areas (Bui et al., 2019; He and Du, 2022; Liu and Zou, 2011; Shen et al., 2023; Tang et al., 2022).

The Theil Index is widely employed in existing literature to quantify the urban-rural development gap. Akita (2017) applied this index within a hierarchical decomposition framework to empirically assess urban-rural expenditure disparities in Indonesia following the 1997 Asian financial crisis. Huang et al. (2020) and Song et al. (2022) utilized the Theil Index to examine impacts on the urban-rural income gap, specifically investigating the roles of infrastructure construction and foreign direct investment, respectively. Guo and Li (2024) measured the urban/rural educational development education inequalities in China by using the Coefficient of variation and the Theil index, finding that the inequalities are reduced in terms of weighted in both of the two indicators. Le and Aralas (2025) also adopted the Theil Index to measure the urban-rural income gap alleviation brought by digital inclusive finance, where there are variations in dimensions and regions. By referring to these empirical researches and according to the definition of Theil Index, it is calculated by:

$$Theil=\mathop{\sum }\limits_{m=1}^{n}\left[\frac{{I}_{m}}{I}\cdot \ln \left(\frac{{I}_{m}/{P}_{m}}{I/P}\right)\right]$$
(4)

Where I and P refer to the total income and the population of the whole region, Im and Pm refer to the total income and the population of region m within. The Theil Index (Theil) turns out to be in the range of 0–1 (can be expressed in percentage (%)), the smaller it is, the narrower the unequal income gap in this region.

In this study, the region of each county is divided into the urban area (m = 1) and the rural area (m = 2), then the Theil index can be identified as:

$$Theil=\frac{{I}_{1}}{I}\cdot \ln \left(\frac{{I}_{1}/{P}_{1}}{I/P}\right)+\frac{{I}_{2}}{I}\cdot \ln \left(\frac{{I}_{2}/{P}_{2}}{I/P}\right)$$
(5)

Empirical analysis model setting

By substituting the core explaining variable (in Eq. 3) and the core explained variable (in Eq. 5) into Eq. (2), the benchmark multi-time-point DID model can be set as:

$$Thei{l}_{it}=\alpha +\beta \cdot DID{m}_{it}+\mathop{\sum }\limits_{j=1}^{n}{\gamma }_{j}Contro{l}_{it}+{\delta }_{i}+{\mu }_{t}+{\varepsilon }_{it}$$
(6)

Where, the explained variable Theilit indicates the income gap in Hubei counties, which ranges from 0 to 100%; the explaining variable DIDmit denotes the execution year and executed county of the GIAP policy, takes 0 or 1. δi signifies the individual fixed effects for the counties, and μt indicates the time fixed effects among them, then εit is the error term. The coefficient β is of most concern: if the estimated \(\hat{\beta }\) is statistically significantly negative, then the GIAP policy is effective and has indeed reduced the rural-urban development divide.

Research data acquisition and description

Followed by Eq. (5), the Theil Index that indicates the rural-urban development divide is composed of the residents’ income Im and population Pm. These two data, along with all the controlled variables in Eq. (6) or listed on rows (3)–(11) in Table 1, are collected from the China Statistical Yearbook (county-level) (2001–2023) and China Rural Statistical Yearbook (2001–2023).

Table 1 Descriptive statistics of research variables.
Full size table

The descriptive statistics of these variables are presented in Table 1, where rows (1) and (2) are the core explaining and explained variables, respectively, rows (3)–(11) display all the controlled variables. All the statistics are panel data that are suitable for the econometrics regression analysis for Model (6).

Results and discussion

Benchmark regression results

To evaluate the effect of GIAP policy on income arising and gap narrowing in urban and rural areas among the counties, the regression results of the benchmark model (Eq. 6) are listed in Table 2.

Table 2 Baseline regression results.
Full size table

In Table 2, column (1) is the analysis result that does not consider control variables and fixed effects. Column (2) includes individual and time fixed effects based on column (1), while columns (3) and (4) add control variables relative to columns (1) and (2). It could be seen that the goodness of fit of the regression (R2) increased sequentially, and whether fixed effects were considered or control variables added, the estimated coefficients of the DID term (\(\hat{\beta }\)) were highly significant at the 1% statistical level. The regression results indicated that the GIAP policy could significantly narrow the level of urban-rural income gap by reducing the Theil Index, which reflected the difference in urban-rural economic level. Quantitatively, the GIAP policy has cut the Theil Index by 4.957 (in column (4)), which is considered to be an optimistic effect. In other words, hypothesis H1.1: The GIAP policy implementation can effectively narrow the urban-rural development gap in Hubei, China, as proved by this empirical evidence.

Robustness test

Parallel trend test and Policy dynamic effect

To ensure the effectiveness of the DID model, a key prerequisite condition is that the treatment group and the controlled group show similar variation tendency in the pre-shock time (when Postit = 0). In the parallel test of benchmark model (6), all the 27 GIAP-tagged counties or districts are regarded as the treated group (colored parts in Fig. 3), whereas other regions without GIAP are the reference group (blank parts in Fig. 3, the districts in the main urban area of the provincial capital city Wuhan and in all other Hubei prefectural-level cities are excluded). If the coefficients of the Theil Index show insignificance before the GIAP execution year and are mostly significant after that, then the parallel test passed. Referring to the existing research (Agüero and Beleche, 2017; Clarke and Tapia-Schythe, 2021; Marcus and Sant’Anna, 2021) and based on the benchmark model (6), an event study model is set to verify the parallel trend as:

$$\begin{array}{ll}Thei{l}_{it}\,=\,\alpha +\mathop{\sum }\limits_{k\geqslant -3}^{-1}\left({\beta }_{k}\cdot {D}_{it}^{k}\right)+{\beta }_{0}\cdot {D}_{it}^{0}+\mathop{\sum }\limits_{k\geqslant 1}^{6}\left({\beta }_{k}\cdot {D}_{it}^{k}\right)\\\qquad\qquad+\mathop{\sum }\limits_{j=1}^{n}{\gamma }_{j}Contro{l}_{it}+{\delta }_{i}+{\mu }_{t}+{\varepsilon }_{it}\end{array}$$
(7)

As those treated counties received the GIAP policy at different time points between 2004 and 2015 (shown in Fig. 4), we shrink the time range into −3 to +6 periods to avoid the multicollinearity in the parallel trend test. In Eq. (7), \({D}_{it}^{k}\) is a dummy variable that represents the year k to (when k < 0) or after (when k > 0) the GIAP execution time. \(\mathop{\sum }\nolimits_{k\geqslant -3}^{-1}\left({\beta }_{k}\cdot {D}_{it}^{k}\right)\), \({\beta }_{0}\cdot {D}_{it}^{0}\) and \(\mathop{\sum }\nolimits_{k\geqslant 1}^{6}\left({\beta }_{k}\cdot {D}_{it}^{k}\right)\) constitute the 3-period pre-policy term, 1-period policy implementation time-point term, and 6-period post-policy term, respectively. The other variables and coefficients are the same as the baseline model Eq. (6). The graphical result of the parallel trend test is plotted in Fig. 5.

Fig. 5
figure 5

Parallel trend test result.

Full size image

As shown in the Fig. 5, the implement of GIAP policy, the estimated coefficients of the pre-policy terms (β−1 to β−3) in the model (7) are all positive but unable to reject null hypothesis (confidence interval include 0); whereas that of the on-policy (β0) and post-policy terms (β1 to β6) are all negative can reject null hypothesis (confidence interval exclude 0). Hence, the parallel trend test results in this study proved that there was an evident and far-reaching effect of the GIAP policy on urban-rural imbalance development.

In fact, the estimation of the on-policy and post-policy terms in Eq. (7) can reflect the year-by-year dynamic effect brought by the GIAP. To quantify this dynamic effect, the event study results are shown in Table 3.

Table 3 Dynamic effect of GIAP.
Full size table

Combined with Fig. 5 and Table 3, the yearly coefficients (D0 to D6) are all significantly negative and are in the range of −5 to −3, proving that the GIAP policy is capable of continuously narrowing the urban-rural development gap at least for 6 years after implementation, though with a slight effect-decreasing as time goes on after the year 1 and 5.

Placebo test

To further enhance the persuasiveness of the DID model, a placebo test is needed to exclude other policy shocks and avoid the unobservable random variables (Heckman and Sedlacek, 1985; JG Slusky, 2017; Lee and Kang, 2006). Specifically, shuffle all variables and randomly assign them to a certain sample to construct a time frame for the GIAP policy to implement. That is, randomly select a certain year as the time for GIAP policy from each county in the experimental group, test the model, and repeat it 500 times to obtain the density distribution map of the estimated coefficient of the stacked sample. As shown in Fig. 6, after 500 simulations, the kernel density distribution map displays a normal distribution around zero, with the horizontal and vertical dashed lines deviating from the density distribution of the random sample coefficient results, indicating that the benchmark regression results mentioned above were not coincidental and did not have major errors, thus demonstrating the irreplaceability of the GIAP policy.

Fig. 6
figure 6

Placebo test result.

Full size image

Further analysis

Heterogeneity analysis

Urban-rural heterogeneity

To directly evaluate the effect of the GIAP policy executed on income arising in urban and rural areas among the counties, the multi-time-point DID models are set as Eqs. (8) and (9):

$$Ainc\_U{r}_{it}=\alpha +\beta \cdot GIA{P}_{i}\times Pos{t}_{it}+\mathop{\sum }\limits_{j=1}^{n}{\gamma }_{j}Contro{l}_{it}+{\delta }_{i}+{\mu }_{t}+{\varepsilon }_{it}$$
(8)
$$Ainc\_R{u}_{it}=\alpha +\beta \cdot GIA{P}_{i}\times Pos{t}_{it}+\mathop{\sum }\limits_{j=1}^{n}{\gamma }_{j}Contro{l}_{it}+{\delta }_{i}+{\mu }_{t}+{\varepsilon }_{it}$$
(9)

Where in Eqs. (8) and (9), the dependent variables Ainc_Urit and Ainc_Ruitrepresent the value of local average income with the implementation of GIAP policy in urban and rural areas at county i in year t, respectively. GIAPi × Postit constitutes the core binary policy variable denoting whether the GIAP policy had been executed on county i in year t, where it takes the value 1 if the county i executed GIAP in year t and takes 0 otherwise. The results are displayed in Table 4.

Table 4 Regression results of average income promotion.
Full size table

Knowing from Table 4, the implementation of the GIAP policy has directly promoted the average income of both urban and rural economies. By comparing the \(\hat{\beta }\) in columns (2) and (4), GIAP’s implementation effect acts actively only in rural areas, as the estimated coefficient is only significant in the rural group. This imbalance in promoting utility may be conducive to the weakening of uneven development between urban and rural areas. That is, the GIAP policy implementation effectively narrows the urban-rural development gap by bringing more incremental per capita income growth to rural areas in Hubei, China. Hypothesis H1.2 is thus proved.

Geographical heterogeneity

Situated in the heart of China and in the middle reaches of the Yangtze River (see Figs. 2 and 3), Hubei province serves as a nationwide hub for logistics and resource transportation. Evidently, the by-river counties can conveniently make use of the Yangtze River shipping, attracting the labor force, funding investment, and industrial agglomeration alongside.

Based on the benchmark model (Eq. 6), we conduct a grouped regression to examine this geographical heterogeneity throughout the GIAP implementation, with the findings presented in Table 5:

Table 5 Regression results of the Yangtze River’s role.
Full size table

To control the geographical influencing factors, we divide the treated groups (affected by GIAP) and the matched groups (without GIAP) into an along-Yangtze River (by-river) group and a far-from-Yangtze River group (far-river), as shown in the row started with “Matched group" in Table 5. From the grouped regression outcome in Table 5, the results are intriguing: (i) By comparing columns (1) and (3) where the matched group are completely the same, the treated counties situated at a greater distance from the Yangtze River can even derive greater advantages from GIAP policy compared to the river-alongside counties; (ii) By comparing columns (2) and (4) where the each matched groups is geographical homogeneous with the treated group respectively, showing the similar estimated coefficient, revealing that the counties adjacent to the Yangtze River, exhibit no significant excess reduction in the urban-rural gap measured by Theil Index. Therefore, Hypothesis H2 is not viable, as it has been disproven by result (i) and is inapplicable to (ii).

Nevertheless, the result (i) does show an exhilarating fact: the GIAP policy can be more beneficial to the counties with poor transportation conditions to some extent. With a view that these regions are relatively underdeveloped and more dependent on primary industries, the identification of local agricultural products brought distinct positive effects in terms of regional synergetic development. The GIAP-tagged commodities originated from the poor-traffic source lands, are usually organic and uncontaminated, having more potential to gain greater market appeal. Moreover, as the market demand for these products raised, the infrastructure construction and the regional coordination works can be in turn facilitated, constituting a virtuous circle for the rural economy.

Policies interaction effects analysis

One of the characteristics of GIAPs is that it is favorable for a country to develop the economy in accordance with local conditions; therefore, the implementation of the GIAP policy is also a vital step in the performance of the poverty alleviation strategy in poor areas. Figure 7 displays the distribution of Hubei counties with GIAP and another supportive policy—the Poverty-Stricken County Identification (PSCI), where the PSCI-identified counties can access the targeted poverty alleviation measures (e.g., fiscal transfer payment, infrastructure construction assistance, and industry support) from the superior government (Yan et al., 2023; Yang et al., 2020; Zhou et al., 2023). As Fig. 7 reveals, PSCI and GIAP policies coexist in many counties that are located on the periphery of Hubei province, China (see the purple parts). The interaction between GIAP and PSCI may have an additive effect on local economic development and urban-rural imbalance reduction.

Fig. 7
figure 7

Distribution of GIAP and poverty counties in Hubei, China.

Full size image

Thus, in this study, policy interaction effects are evaluated by regression with the Theil index as the dependent variable. By incorporating the two-policy interaction term (GIAPi × PSCIi × Postit) and on the basis of Eqs. (3) and (6), we consider the sole effect of PSCI through Eq. (10), and the interaction effect between GIAP and PSCI through Eq. (11):

$$Thei{l}_{it}=\alpha +{\beta }_{2}\cdot PSC{I}_{i}\times Pos{t}_{it}+\mathop{\sum }\limits_{j=1}^{n}{\gamma }_{j}Contro{l}_{it}+{\delta }_{i}+{\mu }_{t}+{\varepsilon }_{it}$$
(10)
$$Thei{l}_{it}=\alpha +{\beta }_{1}\cdot DID{m}_{it}+{\beta }_{3}\cdot GIA{P}_{i}\times PSC{I}_{i}\times Pos{t}_{it}+\mathop{\sum }\limits_{j=1}^{n}{\gamma }_{j}Contro{l}_{it}+{\delta }_{i}+{\mu }_{t}+{\varepsilon }_{it}$$
(11)

Where in Eq. (10), Postit = 1 when individual i was identified with the Poverty-Stricken County at time spot t and otherwise being 0; in Eq. (11), Postit = 1 only when individual i was doubly identified with the GIAP county and Poverty-Stricken County, and otherwise being 0; the other terms are the same with that in Eqs. (3) and (6). The regression results associated with the PSCI policy are outlined in Table 6.

Table 6 Regression results of the interaction of policies of GIAP and PSCI.
Full size table

Table 6 reveals three significative results: (i) Compared \(\hat{{\beta }_{1}}\) (= −2.697) in column (2) in Table 6 and \(\hat{\beta }\) (= −4.025) in the baseline regression (column (4) in Table 4), the former is 49.2% less, revealing that GIAP alone may provide relatively limited gap reduction effect; (ii) Column (4) displays that for those districts and counties without GIAP, the PSCI policy alone can be largely effective; (iii) Distinctly, apart from the effectiveness of the GIAP itself (the statistical significant estimated result of \(\hat{{\beta }_{1}}\), the PSCI policy also positively interact with GIAP, as the estimated interaction term’s coefficient \(\hat{{\beta }_{3}}\) (= −1.670) is significant as well, this indicates that policy mix can play an epoch making role in the county economy. As a result of the PSCI policy, an important targeted poverty alleviation measure is to exploit the procurement market for the rural agricultural products, especially those that are labeled with GIAP. Based on that, Hypothesis H3 is forcefully valid.

Conclusions

During a period of rapid expansion in a developing nation, the urban-rural income gap is usually widened because of unbalanced capital concentration and unidirectional movement of manpower from villages to cities and towns. To compensate for the absence of supportive policies and to promote the value of farm production for the backward rural areas, the GIAP policy has been implemented since the mid-2000s in China, as a part of the plan for the regional coordinated development goal and green ecological agriculture construction.

Measured by the Theil Index with panel data from 2001 to 2023, this paper has examined the GIAP policy implemented throughout 2004–2015 in Hubei, China, aiming at verifying its effectiveness on closing the urban-rural development gap overall and in different natural or social conditions. The empirical verification through the multi-time-spot DID model reflects the conclusions, including (1) On the whole, GIAP has observably diminished the urban-rural gap, resulting in an average of 4.025% reducing effect in the policy-executed counties. The policy dynamic effect test indicates that the treatment effect exists persistently but slightly decreases progressively. (2) The GIAP policy can directly bring about more incremental income for the rural regions but not the urban areas, thereby demonstrating its efficacy in bolstering the agricultural economy. (3) Under the condition of the same matched group, GIAP has brought about 22.1% more urban-rural gap lessening for the counties that are far from the shippable Yangtze River; while when the geographical heterogeneity in the matched groups is controlled homogeneously, the differential effect turns out to be inconspicuous. (4) Combining with another rural revitalization policy—the PSCI, GIAP’s policy impact is remarkably strengthened further.

To conclude, the GIAP policy indeed improved the rural economy and alleviated the urban-rural imbalanced development to a certain degree. Based on the research findings, to maintain the policy sustainability and to deeply actualize the rural revitalization strategies, some policy proposals can be forwarded: (1) GIAP is a continuable policy, more supportive acts, like productive technical support, transportation infrastructure building and agricultural product publicity, are essential to maintain the aforementioned positive effects and to build sustainable rural economies, especially in ecological agriculture areas where transportation is inconvenient. (2) GIAP is also an expandable policy, some other regions that have not GIAP products, can be considered to be the places for expanding the coverage of the GIAP, provided that the environmental assessment and the green productivity fulfill the certification standard. (3) Considering the core element of the GIAP is the geographical feature, the relevant service industries, such as rural ecotourism and e-commerce live-streaming of the GIAP products, can be initiated and progressed to enrich the industrial diversification, bringing about more endogenous growth sources. (4) As the combination of policies GIAP and PSCI proved synergistic, more suitable supportive policies directed at other aspects that current acts have not covered yet, are strongly recommended to work on the rural economy in all directions, especially on the upstream-downstream industries of farm produce. To sum up, promoting the added value of the agricultural products themselves, driving the related industries, and expanding market access and entry may lead to further achievements in the rural revitalization strategy in developing economies.

While our multi-time-point DID model estimates indicate that the GIAP policy alleviates the urban-rural economic gap (measured by the Theil Index), several potential shortages may exist: (1) The Theil Index, as a relative measure, may not fully capture the absolute income levels of the rural population and thus offers an incomplete picture of the urban-rural gap; (2) The study’s geographic scope is confined to a single province, limiting its generalizability to regions with differing cultural and geographic contexts; (3) Potential confounding effects exist, as other rural revitalization policies (e.g., the New Rural Cooperative Medical Scheme, rural e-commerce initiatives) concurrent with PSCI were not controlled for in the empirical analysis.

To address these limitations and advance GIAP research, future studies may be focused on: (i) Conduct policy portfolio analyses using textual data across broader geographic regions; (ii) Employ a wider range of assessment indicators (beyond the Theil Index) to capture multidimensional impacts; (iii) Implement field research on representative agricultural products to assess GIAP’s tangible implementation effects.