What Type of Public Library Best Supports Agricultural Economic Development? An Empirical Study Based on Rural China

Abstract

Modernizing agricultural economies requires the infusion of knowledge and industrialization, necessitating the bridging of the “digital divide” and “talent gap” between urban and rural areas. Public libraries, as key knowledge dissemination institutions, play a crucial role in this process. This study aimed to explore how the development of such institutions can align with agricultural economic growth. Using China as a case study, where the Rural Revitalization Strategy has been implemented in recent years, including the extensive construction of rural public libraries and other infrastructure, we empirically analyzed the correlation between county-level public libraries and agricultural economic development from 2012 to 2019. The results show that the number of county-level public libraries and their assets, collection sizes, e-books, and professional staff have a significant positive impact on agricultural economics, while non-professional staff and facilities have a negative impact. It is recommended that future rural public library development should focus on enhancing professional standards and advancing digitalization and mobile internet integration, while being cautious about expanding the physical scale and staffing. This study fills a gap in the research on the correlation between rural public libraries and agricultural economics, and the methodology employed has a certain degree of general applicability. However, the applicability of the conclusions may be limited by China’s unique national conditions.

1. Introduction

1.1. The Value of Rural Public Libraries and Related Research

The construction of rural public libraries plays a significant role in promoting social, economic, and cultural development in rural areas, with key contributions in the following areas:

• Knowledge Dissemination, Education Enhancement, and Skills Training: Public libraries provide access to a wide range of books and educational resources, which are particularly valuable in rural areas where educational facilities and resources are often lacking. Libraries help residents acquire knowledge and improve their cultural literacy, and they promote lifelong learning. Studies by Eve et al. [1] and Nielsen and Borlund [2] highlight the irreplaceable role of public libraries in lifelong learning, informal education, and continuous vocational skills education;
• Information Access: Modern libraries offer more than just traditional borrowing services; they also provide e-books, online resources, and multimedia content. Many libraries host various training courses and workshops, such as those on agricultural technology, small business management, and computer skills, which are crucial for increasing farmers’ income and employment opportunities. Research by Kinney [3], Abumandour [4], and Manžuch and Macevičiūtė [5] demonstrates that modern digital and networked public libraries are key measures in bridging the digital divide;
• Catalyst for Economic Development: By providing educational resources and skills training, public libraries can enhance the quality of the rural workforce and promote diversified local economic development, helping to mitigate the effects of economic downturns. Moreover, the construction and operation of libraries create job opportunities, both directly and indirectly, contributing to economic growth. Taylor et al. [6] suggest that public libraries help people withstand the effects of economic recessions. Fairbairn and Lipeikaite [7] report that Macedonia’s Braka Miladinovci Public Library provides services for unemployed young women, while South Africa’s Masiphumelele Public Library serves impoverished suburban slums, aiming to help disadvantaged youth embark on positive career paths. In rural areas, Nikcevic-supported services targeting farmers, such as Serbia’s Radislav Nikcevic Public Library, have been instrumental. This library’s AgroLib online market allows farmers to share information on agricultural methods and sell their products. Macedonia’s Goce Delchev-Stip Regional Public and University Library assists farmers in finding subsidy information and applying for grants online, while Lithuania’s Pasvalys Marius Katiliskis Public Library offers ICT training, an online portal, and a desktop publishing center for farmers. Mehra et al. [8] found that rural public libraries have significant potential in driving economic development and sustainable economic vitality in the Appalachian region of the United States;
• Community Centers and Cultural Exchange: Libraries often serve as community hubs, providing a public space for people to gather, interact, and organize events. This contributes positively to community cohesion, cultural preservation, and social integration. Audunson [9] found that public libraries have great potential to become low-intensity meeting places, which play a significant role in promoting tolerance and community integration. Flaherty and Miller [10] found that rural public libraries in the United States have a significant impact on community activities in remote areas, significantly promoting community health work and potentially leading to community transformation in the future;
• Bridging the Digital Divide and Promoting Sustainable Development: As global attention to the Sustainable Development Goals increases, libraries can become platforms for disseminating environmental knowledge and sustainable agricultural practices, helping rural communities address challenges such as climate change and environmental degradation. Audunson et al. [11] demonstrate that public libraries serve as infrastructure for sustainable development, playing a significant role in bridging the digital divide and enhancing cultural inclusiveness. Mansour [12] argues that despite challenges such as weak infrastructure and inadequate collections and related facilities, high illiteracy rates among service recipients, a lack of funding and cooperation between relevant institutions, the improper training of library and information professionals, and a lack of research and surveys, Egypt’s public libraries still contribute positively to sustainable development, although the number is far from sufficient. Panda and Das [13] suggest that public libraries play seven roles in sustainable development: informing, promoting, educating, creating resources, empowering, healing, and advocating. Wang et al. [14] also show that rural public cultural services, represented by public libraries, are crucial for narrowing the rural digital divide, including the physical and capacity divides. Digital public libraries also had unique advantages during the COVID-19 pandemic, with Xin [15] finding that they quickly transitioned their services online in response to the outbreak, actively providing reading therapy, quality digital resources, and reliable pandemic information, offering academic literature support to researchers, and alleviating the psychological pressure of isolated patients and medical staff.

In summary, the construction of public libraries not only enhances the knowledge level and quality of life for rural residents but also provides high-quality talent for rural areas and supports the sustainable development of entire communities. However, not all rural public libraries can achieve significant positive effects. Omeluzor et al. [16] found that rural libraries in Nigeria failed to fulfill their roles. However, scholars also argue that the failure stems from unmet information needs, a lack of up-to-date materials, a lack of awareness, illiteracy, language barriers, insufficient technical staff, and inadequate infrastructure and facilities. It is not just Third-World countries; studies by Bishop et al. [17] and Real and Rose [18] also show that rural public library construction in the United States has many shortcomings.

Given the above, the construction of rural public libraries should be carefully studied to determine how they can best integrate into and adapt to the surrounding social environment based on common economic, political, and social factors and conditions, and how they can be developed to meet the needs of agricultural economic development.

1.2. The Development of Rural Public Libraries in China

In recent years, the Chinese government has placed significant emphasis on rural development, introducing the Rural Revitalization Strategy. The implementation of this strategy relies heavily on the influx of individuals with modern higher education into rural areas. Sun et al. [19] emphasized that, in the context of rural revitalization, the primary focus should be on talent development and a people-centered approach. Wang et al. [20] found in their study on the coupling of the digital economy and rural economy that high-level technological innovation and rural management talent are of paramount importance. Given the significant role of public libraries in talent cultivation, the Chinese government has also placed particular emphasis on the construction of rural public libraries.

China’s public library system has a long history, and, after decades of effort, it has developed into a nationwide network characterized by the following features:

• Tiered Management System: The Chinese government established a public library system quite early, categorized by administrative levels into national, provincial, city, and county libraries, with each level responsible for library services and management within its respective jurisdiction, forming a multi-tiered management and service system [21]. Further down, cultural stations at the township and village levels also provide book-reading services. This system has been gradually perfected over the years and is still in use today [22].

In this tiered system, national-, provincial-, and city-level public libraries are located in the national capital of Beijing, provincial capitals, and prefecture-level cities, respectively, primarily serving urban residents. Although currently open to rural residents, their influence is limited due to geographic constraints. As for the township- and village-level cultural stations, due to uncertain and unstable funding sources, the number of books is very limited, and their impact is also restricted. Only county-level public libraries within the administrative system are closest to rural residents, and a significant proportion of their collections are specialized agricultural books.

A typical county-level public library, such as the one in Dongyang City (a county-level city) under Jinhua Municipality in Zhejiang Province, occupies a building area of 5200 square meters, holds a collection of 200,000 volumes, and is equipped with a book repository, information room, electronic reading room, and children’s reading room, among other facilities. The library staff includes 18 employees, 40% of whom have an associate degree or higher. It accommodates 100,000 to 200,000 visitor instances annually [23]. For context, Dongyang City covers an area of 1747 square kilometers with a permanent population of 1.093 million;

2.

Government Support and Investment: Within the administrative system, national-, provincial-, municipal-, and county-level public libraries receive comprehensive financial support from the respective levels of government, with management personnel incorporated into the civil service. China has also promoted the development of library services through the formulation of relevant policies and regulations. For example, the “Public Library Law of the People’s Republic of China”, promulgated in 2017 [24], clearly outlines the government’s responsibilities and obligations in public library construction and services. In recent years, as China’s economy has grown, the government has increasingly emphasized the construction of a public cultural service system, including the public library system [25].

Note that China is a state-owned country, and the funding for public libraries, including their construction, collection acquisition, operation, and the various activities organized by them, is of a public welfare nature and is supported by state finance. This differs significantly from public libraries in other countries and regions. Especially at the county level, public libraries, mainly serving rural areas and with limited visibility, are almost entirely unable to secure additional funding through private donations.

The advantage of this state-managed mechanism is that the basic funding is guaranteed, eliminating concerns about operational losses for libraries, which allows them to focus on library services. However, the disadvantages are also clear: the system is too rigid and lacks flexibility, making it extremely difficult to scale up or down. Personnel within this system also tend to lose their initiative and settle for the status quo;

3.

Growing Emphasis on Digitalization and Informatization: With the advancement of information technology, China’s public libraries have actively promoted digital transformation. Many libraries now offer e-books, online databases, and digital reading resources while leveraging information technology to enhance their service efficiency and quality. Additionally, due to China’s centralized socialist system, the digital information construction of public libraries is uniformly advanced through administrative directives. For instance, in 2011, the Ministry of Culture of China launched the National Digital Library (NDL) project nationwide [26], significantly enriching the digital resources in the collections of public libraries at all levels within a few years, and making them available for external sharing.

Overall, due to the characteristics of China’s state ownership, the construction of China’s public library system is entirely undertaken by the government, with guaranteed funding sources, allowing for a greater focus on building library infrastructure and enhancing the service quality. However, this government-led public library system can also lead to issues such as rigid management and severe bureaucracy, resulting in inefficiencies. These problems may become more apparent in the development and promotion of rural public libraries and digital libraries.

Shao and colleagues, in their comparative study of public library evaluation systems in China and the United States [27], found that U.S. public libraries place greater emphasis on library services, while Chinese public libraries focus more on infrastructure and basic construction. This is likely because China, as a developing country, still has an underdeveloped public library system. Zhao and colleagues, through a survey of 31 senior library experts across 10 provinces in China, discovered that the development of public cultural services in Chinese libraries is unbalanced and inadequate [28]. They identified external factors such as economic development, government investment in cultural activities, and educational development, along with internal issues like weak management and insufficient investment in cultural services, as the primary influences on the development of public cultural services in libraries. This suggests that for Chinese public libraries to improve their services, they should seek financial support and enhance their internal management. Sun and colleagues [29], through factor and cluster analysis, found that there is a significant disparity in the development levels of public libraries between economically disadvantaged regions and more advanced areas in China, with even economically developed regions like Guangdong facing severe infrastructure deficiencies.

Regarding the construction of rural public libraries and digital public libraries in China, Bin and Miao, in their comparative study on Chinese public libraries and university libraries [30], identified several shortcomings in public libraries, such as the diversification of reader needs, a lack of general electronic publications, low usage skills for electronic publications, the adoption of fee-based services for electronic publications, and weak bargaining and pricing power. Yao and Zhao’s research [31] highlighted deficiencies in some public libraries regarding homepage terminology standardization, unified database search platform selection, and the construction of navigation systems. Wanyan and Hu [32] pointed out that, despite years of development, digital cultural service expenditures remain low among the Chinese population, with significant disparities in digital cultural service participation between different population groups, particularly between urban and rural residents. Although there is strong interest in public digital cultural services, the actual utilization of these services in public libraries is low and fails to meet user quality expectations. Li and Xu [33] argued that while Chinese public libraries play a crucial role in digital resource development and public cultural services, they face contradictions and inconsistencies in areas such as the supply–demand relationship for resources, resource integration, service effectiveness and efficiency, marketing and promotion, and social participation.

1.3. Research Objective

In summary, the current consensus among scholars is that public knowledge and cultural dissemination institutions, such as public libraries, can support sustainable development in rural areas. However, there is still no clear conclusion on how public libraries in rural areas should be constructed to align with the future development of these regions. In recent years, China has vigorously promoted its Rural Revitalization Strategy, which includes the improvement in public cultural service systems and the promotion of public digital libraries. Nonetheless, various concerns remain, as highlighted by scholars’ research.

Moreover, a search using the keywords “public library” and “agriculture” on academic platforms revealed that there is currently very little research that combines public libraries with agricultural economics. In China, studies of this kind are even rarer. As indicated in the Introduction of this paper, research on the impact of rural public libraries on agricultural economics in China is virtually nonexistent. This study aims to fill this gap by exploring the construction direction of rural public libraries that can support sustainable development in rural areas. Given the significant development of rural infrastructure under China’s Rural Revitalization Strategy, this study uses the construction of public libraries in rural China as a case study. By analyzing the correlation between public libraries and agricultural economics across different regions of China, this research seeks to identify the experiences and lessons learned in the construction of rural public libraries, provide policy recommendations for future development, and contribute to the sustainable development of rural areas.

2. Materials and Methods

The research process for this study is outlined in Figure 1 and was as follows:

• Data related to rural public libraries in China from 2012 to 2019 were collected;
• The level and trends of rural economic development in China during the same period were assessed;
• A correlation analysis was conducted based on the aforementioned results;
• The characteristics of rural public libraries that are most beneficial for rural economic development were discussed and summarized.

China comprises 34 provincial-level regions, among which Taiwan, Hong Kong, and Macau were not included in this study due to the significant differences in political systems and statistical standards. Beijing, Shanghai, Tianjin, and Chongqing, the four municipalities directly under the Central Government, predominantly focus on non-agricultural economies. The administrative setup and division of their subordinate units also differ greatly from other provinces and autonomous regions. The setup, collections, and target audiences of the county-level public libraries under their jurisdiction also differ significantly from those in other provinces and autonomous regions and thus were also excluded from this study. The remaining 27 provinces and autonomous regions represent the most significant part of China’s agricultural economy and were sufficiently representative for this analysis.

Data sources: China Statistical Yearbook; China Rural Statistical Yearbook; China Library Yearbook; China Cultural and Heritage Statistics Yearbook; China Culture and Tourism Statistics Yearbook; China Education Statistics Yearbook; among others. In this study, all monetary values in Renminbi were adjusted using historical price indices to reflect constant prices, standardized to the 2019 RMB value, ensuring that the results are more in line with the actual conditions.

This study evaluated the level and trends of China’s agricultural economic development by assessing the production efficiency. It then examined the impact of various data from rural public libraries on the agricultural economic efficiency to explore the future direction of rural public library development. Efficiency evaluation effectively mitigates the disparities in agricultural economies across different provinces and autonomous regions caused by factors such as geography, geology, climate, and historical development. Moreover, efficiency evaluation reflects the sustainability of agricultural economies: a higher efficiency indicates greater output with fewer resource inputs, signifying more advanced agricultural technology, more efficient management, and more rationalized resource allocation. This rationalization is associated with reduced carbon emissions, lower pollutant discharge, and improved sustainability.

Efficiency has long been used as an evaluation metric in economic studies. Schmookler and others used efficiency to evaluate the U.S. economy as early as 1952 [34]. Mundaca et al. employed efficiency evaluation in their study of energy economics [35]. Bukarica and Tomšić used efficiency to assess energy policy and sustainability [36]. In agricultural economics, Paul et al. analyzed U.S. farms and economies of scale using efficiency evaluation [37]. Reza et al. also used efficiency studies to reflect changes in agricultural productivity over 50 years worldwide [38].

This study employed Data Envelopment Analysis (DEA) to assess the production efficiency of China’s agricultural economy. DEA is particularly well suited for evaluating the relative efficiency of decision-making units (DMUs) that involve multiple inputs and outputs. One of its key advantages is that it does not require the allocation of a priori weights to the inputs and outputs, making the analysis more objective and flexible, especially in complex economic and production systems. In this study, the 27 provinces and autonomous regions of China were treated as DMUs, and an input–output-oriented Slack-Based Measure (SBM) model was constructed for the analysis. The non-oriented SBM model can identify and measure the slack in individual input and output variables, providing a more comprehensive and detailed assessment than conventional DEA models.

The formula for the non-oriented SBM model is as follows [39]:

Let the set of DMUs be $j=\{\mathrm{1,2},\dots ,n\}$, with each DMU having $m$ inputs and $s$ outputs. We denote the vectors of the inputs and outputs for ${DMU}_j$ by $x_j={(x_{1j},x_{2j},\dots ,x_{mj})}^T$ and $y_j={(y_{1j},y_{2j},\dots ,y_{sj})}^T$, respectively. We define the input and output matrices (X and Y) as follows:

$$X=(x_1,x_2,\dots ,x_n)\in R^{m\times n}\mathrm{and}Y=\left(y_1,y_2,\dots y_n\right)\in R^{s\times n}.$$

(1)

We assume that all data are positive (i.e., $X>0$ and $Y>0$).

In order to evaluate the relative efficiency of ${DMU}_o=(x_o,y_o)$, we solve the following linear program. This process is repeated $n$ times for $o=(1,\dots ,n)$.

The non-oriented or both-oriented SBM efficiency (${\rho }_{IO}^{*}$) is defined by [SBM-C]:

$$\begin{array}{l}{\rho }_{IO}^{*}=\underset{\lambda ,s^{-},s^{+}}{\min }{\displaystyle \frac{1-(1∕m){\sum }_{i=1}^m(s_i^{-}∕x_{io})}{1+(1/s)\sum _{r=1}^s(s_r^{+}∕y_{ro})}}\\ \mathrm{Subject}\mathrm{to}\\ x_{io}={\displaystyle \sum _{j=1}^n}{x}_{ij}{\lambda }_j+s_i^{-}\left(i=1,\dots ,m\right),\\ y_{ro}={\displaystyle \sum _{j=1}^n}{y}_{rj}{\lambda }_j-s_r^{+}\left(r=1,\dots ,s\right),\\ {\lambda }_j\ge 0\left(\forall j\right),s_j^{-}\ge 0\left(\forall i\right),s_r^{+}\ge \left(\forall r\right).\end{array}$$

(2)

It is particularly important to note that this study employed the global SBM model [40], which treats all DMUs across different periods—namely, the input and output variables of the 27 provinces in China from 2012 to 2019—as a single reference set for model construction. This approach ensures that the efficiency scores obtained account for the influence of different periods. As a result, the efficiency scores and averages of each DMU in each period, as well as the mean efficiency scores of all DMUs in any given period, can be compared longitudinally. The efficiency scores derived from the global SBM model, which incorporates temporal variations, offer a more accurate and realistic fit when performing correlation analysis with rural public library-related data that also vary over time. Additionally, by aggregating the average efficiency scores of all DMUs across different periods, the global SBM model can be used to observe the development trends in the overall economic efficiency of China’s agriculture.

In the correlation analysis, this study employed a regression model. It is important to note that the data related to rural public libraries used as independent variables in this study can only partially explain the dependent variable, which is the agricultural economic efficiency score, indicating a limitation of the model. However, the primary objective of this study was to identify the characteristics of public libraries that are most conducive to agricultural economic development and to clarify the future direction of rural library development. Therefore, the focus was on establishing the correlation between rural public libraries and agricultural economics rather than on determining causality, making the regression model a suitable method for this research.

This study specifically utilized a Tobit regression model because the dependent variable was the DEA efficiency score of agricultural economics, which is greater than 0 and less than or equal to 1. The Tobit regression model is designed to account for such censored dependent variables, using the maximum likelihood estimation method to effectively prevent bias in the model’s estimates. It separates the effects of the censored and uncensored portions of the dependent variable, ultimately converting it into a probability model, making it more suitable for models with complex data characteristics. Moreover, the maximum likelihood method also offers a higher tolerance for multicollinearity among the variables compared to other methods.

3. Results

The input and output variables used in constructing the global SBM model for the agricultural economic efficiency across 27 provinces in China from 2012 to 2019 are presented in

Table 1. Input and output variables of the global SBM model for China’s agricultural circular economy.

Indicator Categories	Indicators
Input Indicators	Rural Population
	Consumption of Chemical Fertilizers
	Consumption of Pesticides
	Consumption of Diesel Fuel
Output Indicators	Agriculture, Forestry, Animal Husbandry, and Fishery Total Output Value
	Primary Industry Added Value

. The panel data for these input and output variables across the 27 provinces from 2012 to 2019 can be found in Supplementary Table S1.

It is important to note that certain input variables commonly considered significant, such as land area and irrigated area, were not included in this study. This decision was made due to the significant variations across the 27 provinces. For instance, in the northern regions, crops are harvested once a year, whereas in the southern regions, they may be harvested three times annually, leading to substantial differences in the output per unit area. Additionally, provinces like Inner Mongolia and Ningxia focus primarily on animal husbandry, while southern provinces have a larger proportion of their agricultural activities devoted to crop cultivation. Agricultural production in the northern and southern regions was specifically adjusted to account for these differences.

The input variables selected for this study are less affected by such regional disparities. For example, the “Consumption of Diesel Fuel” is used in both crop cultivation and animal husbandry, and its consumption is independent of the number of harvest cycles per year. Furthermore, these variables constitute a significant portion of the overall costs. Other inputs, such as seeds and plastic film, which have a smaller cost share and limited usage, were not included in this study, as their impact on the model results would have been minimal and might have introduced inaccuracies.

Regarding the output variables, the “Agriculture, Forestry, Animal Husbandry, and Fishery Total Output Value” represents the annual total output value of the agricultural economy across the 27 provinces, while the “Primary Industry Added Value” can be considered the gross profit of the agricultural economy in these provinces.

The DEA analysis in this study was conducted using DEARUN software, version V3.2.0.2. This software features a user-friendly graphical interface with built-in DEA model algorithms, eliminating the need for users to write custom code. The average agricultural economic efficiency scores for the 27 provinces and autonomous regions from 2012 to 2019 are shown in

Table 2. Mean scores of the global SBM model for the agricultural circular economy in 27 provinces and autonomous regions of China, 2012–2019.

	TE	PTE	SE
2012	0.500286	0.599746	0.863904
2013	0.521391	0.635033	0.852163
2014	0.52549	0.636519	0.852553
2015	0.534844	0.659064	0.843958
2016	0.566399	0.690096	0.84505
2017	0.5541	0.653512	0.867062
2018	0.594693	0.714161	0.852686
2019	0.68224	0.83108	0.833445

. TE represents the average technical efficiency score under the assumption of constant returns to scale, PTE represents the average pure technical efficiency under variable returns to scale, and SE represents the average scale efficiency score. Notably, the TE is the product of the PTE and SE. Detailed data from the model can be found in Supplementary Table S2.

In selecting the independent variables for the Tobit model, this study first excluded national, provincial, and municipal public libraries in China, as these libraries are located in the capital, provincial capitals, or major regional cities and primarily serve urban residents. This study focused on county-level public libraries, as these libraries primarily serve the rural population.

The statistical data for China’s county-level libraries, varying by year, include over 40 items. Based on the importance and completeness of the data, this study ultimately selected 13 variables related to county-level public libraries in 27 provinces and autonomous regions in China as the independent variables for the Tobit regression model:

• The number of county-level public libraries, number of staff, number of professional staff, and total assets of county-level libraries, representing the actual scale of the county-level libraries in each province and autonomous region;
• The revenue of county-level public libraries, representing government support, as the vast majority of county-level library revenue comes from government appropriations;
• The number of books, the number of library cards issued, the building area, and the number of reading room seats, representing the professional development of county-level libraries;
• The number of e-books and the number of public computers, representing the level of digitization of county-level libraries and the role of public libraries in bridging the “digital divide”;
• The circulation number of library visits and the turnover rate of the collection, representing the operation and activity level of county-level public libraries.

Considering the differences between the provinces and autonomous regions, these values were normalized by dividing by the rural population of each province and autonomous region, resulting in 13 independent variables. Details of these variables are provided in Supplementary Table S3.

This study used the SPSSAU online analysis software to construct the Tobit model, with version number SPSSAU24.0. This software does not require downloading and comes with a variety of built-in data analysis model algorithms, allowing users to directly input data and construct models without the need to write model code. Additionally, it provides some basic analytical functions.

Table 3. Likelihood ratio test for Tobit Model 1 with constant returns to scale.

Model	−2× Log-Likelihood	Cardinality	df	p	AIC	BIC
Intercept Distance	−122.651				−18.686
Final Model	−240.775	118.123	13	0.000	−212.775	−165.521

presents the likelihood ratio test results for the Tobit model (hereafter referred to as Model 1), with the TE scores as the dependent variable and 13 county-level public library data variables as the independent variables. The p-value is less than 0.05, indicating that Model 1 rejects the null hypothesis. The Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) were used to assess the goodness of fit for the model’s data, with smaller values indicating a better fit. Based on these values, Model 1 demonstrates a good fit.

Table 4. Results of Model 1 with constant returns to scale.

	Regression Coefficient	Standard Error	z-Value	p-Value	95% CI
Intercept	0.493	0.031	15.803	0.000	0.432~0.554
County Libraries per 10,000 Rural Residents	1.363	0.242	5.640	0.000	0.889~1.837
County Library Staff per 10,000 Rural Residents	−0.750	0.124	−6.069	0.000	−0.992~−0.508
County Library Professionals per 10,000 Rural Residents	0.618	0.157	3.948	0.000	0.311~0.925
County Library Collections per 10,000 Rural Residents	0.000	0.000	2.954	0.003	0.000~0.000
E-Books in County Libraries per 10,000 Rural Residents	0.000	0.000	4.475	0.000	0.000~0.000
Public Computers in County Libraries per 10,000 Rural Residents	0.016	0.027	0.588	0.557	−0.037~0.069
Library Cards Issued by County Libraries per 10,000 Rural Residents	0.000	0.000	1.721	0.085	−0.000~0.000
Circulation of County Libraries per 10,000 Rural Residents	−0.000	0.000	−2.497	0.013	−0.000~−0.000
Books Borrowed from County Libraries per 10,000 Rural Residents	−0.000	0.000	−0.792	0.428	−0.000~0.000
County Library Revenue per 10,000 Rural Residents	0.005	0.004	1.507	0.132	−0.002~0.013
County Library Assets per 10,000 Rural Residents	0.000	0.000	2.344	0.019	0.000~0.001
County Library Area per 10,000 Rural Residents	−0.002	0.001	−3.348	0.001	−0.003~−0.001
County Library Seats per 10,000 Rural Residents	0.016	0.006	2.510	0.012	0.003~0.028
Log (Sigma)	−1.976	0.048	−41.076	0.000	−2.071~−1.882

Dependent variable: TE; McFadden R²: −0.963.

presents the results of Model 1, which was constructed using TE scores as the dependent variable and 13 county-level public library data variables as the independent variables.

Table 5. Likelihood ratio test for Tobit Model 2 with variable returns to scale.

Model	−2× Log-Likelihood	Cardinality	df	p	AIC	BIC
Intercept Distance	−24.800				−18.686
Final Model	−101.696	76.895	13	0.000	−73.696	−24.442

presents the likelihood ratio test results for the Tobit model (hereafter referred to as Model 2), constructed by replacing the dependent variable with PTE scores. The p-value is also less than 0.05, indicating that Model 2 similarly rejects the null hypothesis. The AIC and BIC values are slightly higher compared to those of Model 1.

Table 6. Results of Model 2 with variable returns to scale.

	Regression Coefficient	Standard Error	z-Value	p-Value	95% CI
Intercept	0.635	0.043	14.758	0.000	0.550~0.719
County Libraries per 10,000 Rural Residents	1.782	0.334	5.344	0.000	1.128~2.436
County Library Staff per 10,000 Rural Residents	−0.884	0.170	−5.187	0.000	−1.218~−0.550
County Library Professionals per 10,000 Rural Residents	0.975	0.216	4.511	0.000	0.551~1.398
County Library Collections per 10,000 Rural Residents	0.000	0.000	4.119	0.000	0.000~0.000
E-Books in County Libraries per 10,000 Rural Residents	0.000	0.000	1.439	0.150	−0.000~0.000
Public Computers in County Libraries per 10,000 Rural Residents	−0.099	0.037	−2.648	0.008	−0.172~−0.026
Library Cards Issued by County Libraries per 10,000 Rural Residents	0.000	0.000	2.197	0.028	0.000~0.000
Circulation of County Libraries per 10,000 Rural Residents	−0.000	0.000	−1.842	0.065	−0.000~0.000
Books Borrowed from County Libraries per 10,000 Rural Residents	−0.000	0.000	−1.321	0.186	−0.000~0.000
County Library Revenue per 10,000 Rural Residents	−0.005	0.005	−0.970	0.332	−0.015~0.005
County Library Assets per 10,000 Rural Residents	0.000	0.000	1.575	0.115	−0.000~0.001
County Library Area per 10,000 Rural Residents	−0.003	0.001	−3.191	0.001	−0.005~−0.001
County Library Seats per 10,000 Rural Residents	0.026	0.009	3.016	0.003	0.009~0.044
Log (Sigma)	−1.654	0.048	−34.385	0.000	−1.749~−1.560

Dependent variable: PTE; McFadden R²: −3.101.

presents the results of Model 2, which was constructed by replacing the dependent variable TE scores with PTE scores.

4. Discussion

The efficiency of China’s agricultural economy during the period covered by this study is presented in Table 2. This study employed a global SBM model, which allowed for cross-period comparisons, and the data from Table 2 are visualized in Figure 2, which includes a linear trendline for a clearer interpretation. As shown in Figure 2, the TE scores of China’s agricultural economy steadily improved from 2012 to 2019, with only a slight decline observed between 2016 and 2017, followed by a recovery that not only compensated for the previous dip but also surpassed the prior levels. Notably, 2018 marks the initiation of China’s Rural Revitalization Strategy, suggesting that the comprehensive policies under this strategy, including the substantial development of rural public libraries, have been effective at enhancing the efficiency of China’s agricultural economy.

As previously mentioned in the Results Section, the TE is the product of the PTE and SE. Figure 2 also illustrates that the primary factor contributing to China’s agricultural economic efficiency, as indicated by the TE score, is the SE, while the PTE generally remains lower. Scale efficiency evaluates whether increases in the input and output variables yield proportional returns, whereas the PTE encompasses factors beyond the scale, such as agricultural technology levels, production management, financing capabilities, and marketing. The results shown in Figure 2 suggest that relative to the scale of the industry, China’s agricultural economy is weaker in areas related to pure technical efficiency.

However, the trend lines for the three scores reveal that the upward trend in the agricultural economic efficiency is driven by continuous improvements in the PTE. For the study period, the trend for the SE is declining, while the trend for the PTE shows a steep upward slope, closely aligning with the changes in the TE score. This indicates that the improvement in the pure technical efficiency has become the key factor driving the overall increase in the agricultural economic efficiency. The focus of this study on the construction of rural public libraries—aiming to disseminate agricultural knowledge in rural areas and bridge the “talent gap”—clearly falls within the scope of factors contributing to the pure technical efficiency in the agricultural economy.

This study constructed two Tobit models with different dependent variables, a common approach to testing model robustness. Xu et al. [41] in their study on the impact of urbanization on rural land transfers, Su et al. [42] in their research on the digital economy, and Lin et al. [43] in their investigation of the impact of air pollution on technological innovation all employed the strategy of replacing the dependent variable to test the robustness of their models. The high similarity in the results of the two Tobit models in this study thus confirms the robustness of the models.

As shown in Table 4, Model 1, which uses TE scores as the dependent variable, six out of the thirteen independent variables related to county libraries have a significant positive impact on the agricultural economic efficiency TE scores under the assumption of the CRS. These variables are as follows: County Libraries per 10,000 Rural Residents; County Library Professionals per 10,000 Rural Residents; County Library Collections per 10,000 Rural Residents; E-Books in County Libraries per 10,000 Rural Residents; County Library Assets per 10,000 Rural Residents; and County Library Seats per 10,000 Rural Residents. Three variables have a significant negative impact on the TE scores: County Library Staff per 10,000 Rural Residents, Circulation of County Libraries per 10,000 Rural Residents, and County Library Area per 10,000 Rural Residents.

As shown in Table 6, Model 2, which uses PTE scores as the dependent variable, five out of the thirteen independent variables related to county libraries have a significant positive impact on the agricultural economic efficiency PTE scores under the assumption of the VRS. These variables are as follows: County Libraries per 10,000 Rural Residents; County Library Professionals per 10,000 Rural Residents; County Library Collections per 10,000 Rural Residents; Library Cards Issued by County Libraries per 10,000 Rural Residents; and County Library Seats per 10,000 Rural Residents. Three variables have a significant negative impact on the PTE scores: County Library Staff per 10,000 Rural Residents; Public Computers in County Libraries per 10,000 Rural Residents; and County Library Area per 10,000 Rural Residents.

Based on the combined results of the two Tobit models, the number of county-level public library-related variables that have a positive impact on the agricultural economic efficiency is significantly greater than those with a negative impact. This indicates that the construction and development of county-level public libraries in China have an overall positive effect on the agricultural economic efficiency. Numerous studies support the positive influence of rural public libraries on agricultural development, such as Bini Okiy’s research on rural public libraries in Nigeria [44], Sultana et al.’s study on rural public libraries in West Bengal [45], as well as studies in developed countries like Svendsen et al.’s research on rural public libraries in Denmark [46] and Sikes’ study on rural public libraries in Washington County [47].

County Libraries per 10,000 Rural Residents, County Library Assets per 10,000 Rural Residents, County Library Collections per 10,000 Rural Residents, and County Library Seats per 10,000 Rural Residents all demonstrate a significant positive impact on China’s agricultural economic efficiency in both Tobit models. In contrast, County Library Area per 10,000 Rural Residents shows a significant negative impact on the efficiency of the cultural industry in both models. These findings suggest that while the number and total assets of county-level public libraries should continue to increase, resources should not be allocated to expanding physical space but rather to enhancing the collection of books and providing more opportunities for on-site reading. Sultana’s research on rural public libraries in West Bengal [45] similarly found that library collections and related metrics are more crucial. Mansour et al.’s study of public libraries in Egypt [12] also indicates that infrastructure and collections are critical to the quality of public library services.

It is important to note that the independent variables used in the Tobit model of this study represent rural public library resources per 10,000 rural residents in China. However, due to ongoing urbanization, the rural population in China has been steadily declining, with the proportion of rural residents decreasing from 47.43% in 2012 to 39.40% in 2019, and the total rural population decreasing from 642.22 million in 2012 to 551.62 million in 2019 [48]. This trend implies that even without directly increasing the number of rural library resources, merely maintaining the current scale would lead to an annual increase in public library resources per 10,000 residents. Therefore, this further emphasizes that when the government strengthens the construction of rural public libraries, the focus should be on optimizing the structure of existing investments rather than on simply increasing resources such as building space.

The number of county library staff per 10,000 rural residents has a significant negative impact on the agricultural efficiency in China in both models. Conversely, the number of county library professionals per 10,000 rural residents has a significant positive impact in both models. This suggests that the number of non-professional staff and even the total number of staff in rural public libraries should be reduced, while the number of professional staff should be increased, indicating a need for greater professionalization in rural public libraries. As China is a state-owned country, the primary goal of government-managed public libraries is not profit making, funding sources are very stable, and staff management is within a fixed establishment. This model is likely to lead to rigid management and overstaffing, especially as non-professional staff may become a burden in the daily operation of public libraries. An excessive number of such staff can directly consume library resources and may lead to decreased work efficiency and quality. Future construction of rural libraries should particularly focus on improving the efficiency of human resources. De Jager et al. [49] also found in their study on South African rural public libraries that the quality of the staff significantly impacts the quality of library services. Rehman et al. [50], in their study on human resource management in public organizations in Pakistan, mentioned the same issue: non-professional staff within the establishment can severely affect the service quality. Research by Real and Rose [18] on remote rural public libraries in the United States found that the level of service, programming, and full-time equivalence of staff are important for effectively enhancing the library service quality. Mansour [12] also shows that the professionalization and informatization levels of library staff are very important.

The variable “E-Books in County Libraries” has a significant positive impact in Model 1, where the TE efficiency scores are the dependent variable, but shows no significant impact in Model 2, where the PTE efficiency scores are the dependent variable. Since the TE is the product of the PTE and SE, this indicates that the influence of “E-Books in County Libraries” on the agricultural economic efficiency is primarily reflected in the improvement in the scale efficiency (SE). In contrast, “Public Computers in County Libraries” has a significant negative impact on the PTE efficiency scores, which may seem contradictory to the performance of the “E-Books in County Libraries” variable. However, it is important to note that e-books represent software, which can be accessed and read on various hardware devices, such as computers and mobile phones, while “Public Computers in County Libraries” represents hardware. Thus, considering these factors together, the future development of rural libraries should emphasize digitalization and mobile internet capabilities, focusing on increasing the availability of software resources, such as e-books, while appropriately reducing hardware facilities like computers to conserve resources.

It should be noted that the Chinese Copyright Law lists 12 scenarios for the fair use of intellectual property, of which the following are applicable to library and information institutions: (1) for individual learning, research, or appreciation, use of another’s published works; (2) for the introduction, review of a work, or explanation of an issue, appropriately quoting another’s published works within a work; (3) for the translation or limited reproduction of published works for classroom use or scientific research in schools, for use by teaching or research staff, but not for publication or distribution; (4) for libraries, archives, memorials, museums, art galleries, etc., to copy works in their collections for display or preservation purposes; (5) to sketch, paint, photograph, or video art works displayed in outdoor public places.

Based on this, in China, various types of libraries have achieved the online sharing of e-books to some extent. These libraries typically obtain e-books through legal channels, including purchasing licenses from publishers or using platforms that provide authorized access to e-books to manage copyright issues. These licenses allow libraries to lend digital copies to their patrons under specific conditions that respect copyright law or to offer online reading. Additionally, some libraries may participate in digital consortia, sharing resources among several institutions to further expand access while adhering to copyright agreements. Specifically for county-level public libraries, online shared e-books are of a public welfare nature, and the related copyright fees are paid by the libraries themselves, funded by the “Special Construction Funds for Digital Libraries” supported by finance. How these circulation methods impact the copyright of e-books is beyond the scope of this study.

E-books, characterized by rapid dissemination and appeal to younger audiences, are crucial for quickly bridging the “digital divide” and “talent gap” between urban and rural areas and addressing educational inequities between them. Real and Rose [18] found in their study on U.S. rural public libraries that wireless public internet access and the broadband capacity significantly impact the quality of rural library services, underscoring the importance of electronic information dissemination via the internet as a critical responsibility of contemporary public libraries, especially those in remote areas. Santoso et al.’s research suggests that e-books are more effective than physical books and are vital for improving education in rural schools in Indonesia [51]. Zhang and Zhou [52] argue that Education Informatization 2.0, which includes e-books, places a greater emphasis on educational equity. Additionally, Prasetianto et al.’s research indicates that digital reading can significantly enhance agronomy students’ reading comprehension in an English-language environment [53]. In conclusion, the increase in e-books within rural public libraries plays a significant role in disseminating specialized knowledge to rural residents, thereby contributing to the development of more skilled professionals. These newly developed talents are crucial for restructuring the industry and enhancing the scale efficiency.

It is also noteworthy that the Circulation of County Libraries per 10,000 Rural Residents has a significant negative impact in Model 1, contrary to intuition, which generally suggests that the circulation of collections truly represents their value. It should be explained that, according to the China Library Yearbook, the circulation statistics only include printed materials and physical audiovisual documents (such as CDs), excluding the circulation of e-books. In today’s era of the rapid development of mobile internet, printed materials and physical documents carried by CDs are being largely replaced by online-distributed e-books, which explains the annual decline in circulation. Correspondingly, there is an annual improvement in the agricultural economic efficiency, so the negative impact on the agricultural economic efficiency reflected in the model is normal. This also inadvertently underscores the importance of e-books and the construction of digital libraries from another perspective. This simultaneously highlights a regret: due to the lack of circulation data for e-books, the study lacks direct evidence of their impact.

5. Conclusions

In this empirical analysis using China as a case study, the following conclusions were drawn:

Between 2012 and 2019, the efficiency of China’s agricultural economy showed a continuous improvement. Correlation analysis indicates that the number of rural libraries per capita has a positive impact on China’s agricultural economy. Therefore, it is recommended to increase the number of rural public libraries per capita in rural areas.

During the period from 2012 to 2019, the agricultural economic efficiency in China progressively increased. Correlation analysis indicates that the number of rural libraries per capita among rural residents has a positive impact on China’s agricultural economy. Currently, it is advisable to increase the number of rural public libraries per capita. Given the current situation in China, where urbanization is leading to a rapid decline in the rural population, in most cases, maintaining the current number of rural public libraries is sufficient to effectively increase the per capita availability.

The construction of rural public libraries should focus on enhancing their software capabilities and improving their service levels for readers. Specific measures include increasing the number of books, expanding the collection of e-books, and providing more reading seats for patrons. This implies that, initially, it is essential to enhance the management efficiency of libraries. Subsequently, for the same reasons, maintaining or slightly increasing the number of professional staff can achieve the same effect.

The hardware facilities of rural public libraries should be appropriately reduced, such as discontinuing the further construction of buildings and reducing the number of computers, among other measures. The funds saved should be allocated to the construction and maintenance of digital libraries. Special attention should be given to new public library constructions, particularly in underdeveloped areas with insufficient infrastructure.

In summary, this study suggests that future rural public libraries should become more professional, digital, and mobile internet-based. The future direction of development should focus on providing more professional services to rural residents through more robust information technology.

It is important to note that China’s unique characteristics, such as its highly centralized government, may limit the applicability of these findings to other regions. However, the combined research methodology of using a global DEA model with a Tobit model, as applied in this study, possesses a degree of generalizability.

Regarding future developments of this study, plans are in place to address its current limitations. Initially, this research merely touched upon the correlation between rural public libraries and agricultural economic efficiency. Future efforts will attempt to delve into the causality between the two. Furthermore, further detailed investigations are planned, such as examining the impacts of various factors, including the types of collections in public libraries, their specialties, and publication years, as well as the influence of e-book circulation volumes and distribution methods.