What Drives the Development and Sustainable Growth of Cultural Nonprofits—Chinese Province-Level Evidence

Abstract

Although cultural nonprofits play an increasingly important role in the provision of public cultural services in China, there are obvious regional differences in the development of cultural nonprofits. What factors affect this regional difference? This paper builds a theoretical model to explain the regional differences from the perspectives of regional demand, resource supply, and agglomeration effects. Data from the 31 provinces in mainland China from 2010 to 2015 are used to empirically examine the research model. The results indicate that demand for heterogeneity, financial resources, and human resources have positive effects on the size of cultural social organizations, and that there are also significant agglomeration effects with respect to the sustainable growth of cultural nonprofits; however, these findings vary across types of Chinese nonprofits (social organizations, private non-enterprise organizations, and foundations). These findings improve our understanding of regional differences of Chinese cultural nonprofits and have important policy implications for governments to promote the development of cultural nonprofits.

1. Introduction

Since China’s reform and opening-up, nonprofit organizations have begun to enter the public cultural field. With the rapid development of the economy and the improvement in people’s living standards, citizens have sought more culturally enriched lifestyles and have exhibited an interest in cultural activities and programs. Accordingly, differences in people’s social classes, occupations, education levels and economic conditions have led to multilevel and diversified cultural demands. The single public cultural service model supported by government finances and provided by cultural institutions has been unable to meet these diverse demands. To alleviate the contradiction between the supply of and demand for public culture, the Chinese government has proposed cultivating cultural nonprofit organizations to participate in the construction of a public cultural service system. As a consequence, cultural nonprofit organizations have developed rapidly and have begun to provide various public cultural services; however, what factors influence the regional development of cultural nonprofits is still poorly understood. The answer to this question has an important practical significance for understanding the developmental mechanisms of Chinese cultural nonprofits and for contributing to the development of the means by which governments can promote their development.

The regional development of nonprofit organizations has increasingly attracted the attention of researchers who have found significant differences in the size of the nonprofit sector at the global macroregional level and at the regional level within national states. Various theories have been developed to explain the factors that determine the development of the nonprofit sector, including government failure theory [1], interdependence theory [2], entrepreneurship theories [3], and social capital theory [4]. Based on these theories, scholars have conducted a large number of empirical analyses in different countries, including the United States [4,5], Spain [6], the Netherlands [7], Brazil [8], and China [9], as well as in different fields, such as education, health, social services, women- and children-focused services, and poverty alleviation [8,10]. In China, although research on the regional development of nonprofit organizations has attracted scholarly attention in recent years, current scholarship lacks studies specifically focused on cultural nonprofits, whereas empirical studies have shown that the factors determined the development of nonprofit organizations vary across nonprofit fields [8,11]; therefore, it is necessary to conduct an in-depth analysis of the influence mechanism of the development of cultural nonprofits.

The aim of this study is to understand the determinants of the regional development of cultural nonprofits at the provincial level in China. Based on a literature review, we develop a theoretical framework from three perspectives: regional demand, resource supply, and agglomeration effects. In addition, panel data from 31 provinces from 2010 to 2015 were used to test the research hypotheses. The results reveal that regional demand, financial resources, and human resources have important effects on the size of cultural nonprofit organizations. Moreover, there are significant agglomeration effects during the growth of cultural nonprofits. Further analysis indicates that these findings vary across types of Chinese nonprofits.

2. Theoretical Framework

2.1. Regional Demand

Dominant demand theory is market failure and government failure theory. Market failure theory argues that due to the nature of pursuing profit maximization, enterprises lack the motivation to provide noncompetitive and nonexclusive public services, and are unwilling to serve specific people who lack purchasing ability, thus resulting in market failure. Although government and nonprofit organizations are better providers of these services, due to financial and capacity constraints, the government cannot guarantee adequate provision of the total amount of public services, and therefore, it pays more attention to the preferences of the median voters. In this case, the individualized demand of different groups for public services cannot be satisfied, which leads to government failure [1]. In China, the supply of public culture services also faces the problem of government failure. Since the reform and opening-up of China, the economy has grown rapidly, and people’s cultural needs have become increasingly diversified; however, some vulnerable groups, such as migrant workers, left behind children, and nesters cannot effectively influence the government decision-making process for public cultural services due to the lack of an institutionalized channel for expressing their interests, which results in an inability to effectively satisfy their demands for public culture [12]. Accordingly, market failure and government failure provide the necessary development space for nonprofit organizations as they combine the flexibility of the market and the public welfare interest of the government, and can provide a variety of public cultural services according to the varied needs of the people. The non-distribution constraint also determines that nonprofit organizations are more suitable than the government or the market to provide public cultural services for special groups; therefore, nonprofit organizations become a beneficial supplement to the market and the government in a public cultural service system.

Although a core concept of the theory of government failure is demand heterogeneity, Weisbrod [1] argued that meeting diversified demands is an advantage of nonprofit organizations over markets and governments. More specifically, the more diverse the cultural needs of a region’s citizens are, the smaller the market for certain services. Hence, as enterprises lack the incentive to provide such services, and governments cannot meet the needs of different groups, the size of regional cultural nonprofit organizations depends on the heterogeneity of regional cultural needs and the extent to which government departments meet the diverse cultural needs of residents. Prior studies have measured regional demand heterogeneity by race heterogeneity [5,13], religious heterogeneity [14,15], age heterogeneity [16], and education heterogeneity [17]. In China, it is not appropriate to use religious diversity or ethnic diversity to measure the heterogeneity of regional demand. Thus, this study analyzes the development of nonprofit organizations based on age heterogeneity and education heterogeneity. The research suggests that age influences people’s participation in public cultural activities. For example, Courty and Zhang [18] used survey data from 13 major Chinese cities and found that there are complex generational effects regarding the participation of different age groups in public cultural activities [18], with the demands for public cultural services varying by age group; therefore, age diversification leads to improvements in the heterogeneity of regional public cultural needs. Education, which is an important indicator for measuring people’s social and economic status, plays an important role in modern society and can promote and serve as an important factor in social stratification [19]. That said, people with different levels of education have different demands for public cultural services due to differences in lifestyle, cultural beliefs, and so on; therefore, diversity among the levels of education in a region increases the diversity of people’s cultural needs. Based on the above analysis, the following hypotheses are proposed:

Hypothesis (H1).

The greater the age heterogeneity in a region is, the larger the size of cultural nonprofit organizations.

Hypothesis (H2).

The greater the education heterogeneity in a region is, the larger the size of cultural nonprofit organizations.

From the perspective of demand, vulnerable groups, such as the unemployed, the elderly, children, and people who are illiterate, place greater demands on cultural nonprofit organizations. On the one hand, the lack of social resources for vulnerable groups makes enterprises unwilling to reduce profits to provide corresponding cultural services for these vulnerable groups for whom culture becomes a luxury [20]. On the other hand, the weak position of these groups in politics leads to the failure of the government to supply public cultural services [1]; therefore, cultural nonprofit organizations become the main force for addressing the cultural needs of these at-risk groups. According to organizational ecology theory, at the regional level, the higher the demand of vulnerable groups for public cultural services is, the more willing they are to accept the social services through nonprofit organizations, and the greater the environmental capacity of the region for nonprofit organizations [21]. Consequently, more cultural nonprofit organizations exist to provide services in such regions to meet the cultural needs of groups. The existing studies have measured the needs of vulnerable groups in regions through population proportion indicators such as the impoverished population [22,23], the unemployed population [13], the elderly population [24], and minority populations [16,23]. Considering the reality of China and the availability of data, this study used the dependency ratio of the elderly population, the proportion of the population that is illiterate, and the unemployment ratio to measure the degree of socially vulnerable groups in a region.

At present, China has a rapidly aging society, and the change in the older Chinese population (i.e., more than 60 years of age), which has increased from 13.26% in 2010 to 16.15% in 2015, reflects the considerable and rapid growth in this population. Moreover, there are substantial regional differences in this population [25]. Pension services for the aging population have become a major problem in Chinese society, but in the face of substantial demand, they are even more insufficient in quantity and low in quality. By the end of 2010, China’s pension beds covered only 1.8% of the elderly population, and there was a serious discrepancy between the supply and demand of pension services. In this context, the state council and local governments issued relevant policy documents, whereby, with respect to policies and funds, the government supports the participation of nonprofit organizations in providing pension services for the elderly to compensate for the deficiencies of the government, the market, and families to provide adequate pension services [26]. To meet the spiritual and emotional needs of the elderly, “culturally support the aged” (in a narrow sense, “culturally support the aged” refers to meeting the spiritual needs of the elderly after retirement through various cultural service activities) is a new type of pension model that has been explored in recent years. Given such social needs, cultural nonprofit organizations can capitalize on their own advantages and implement various targeted cultural services projects according to the characteristics of the varied needs of the elderly groups. This paper uses the dependency ratio of the elderly population to reflect the burden of an aging society in a region. We suggest that the greater the burden of supporting the elderly population in a region is, the greater the demand for nonprofit organization services designed to “culturally support the aged”, and the greater the size of cultural nonprofit organizations in the region.

Illiteracy and unemployment are often linked to poverty. Due to the lack of relevant knowledge and skills, individuals who are illiterate or unemployed have difficulty finding jobs, often live in poverty, and require society to provide them with their basic needs; however, poverty alleviation should not only be supported economically and materially, but also culturally and spiritually, to help poor people out of poverty by changing their spiritual outlook. Chinese nonprofit organizations have conducted considerable work in the field of cultural poverty alleviation over a long period of time. Under the current strategy of targeted poverty alleviation in China (https://en.wikipedia.org/wiki/Targeted_Poverty_Alleviation, accessed on 1 May 2020), nonprofit organizations have become an important social force in the field of poverty alleviation by forming a three-way pattern of cooperation with the government and the market. The concept of rooted in the grass-roots units of nonprofit organizations, meets the strategic needs of the targeted poverty alleviation”, and provides diversified public cultural services to the poor; therefore, we predict that the more illiterate and unemployed people there are in a region, the greater the promotion of the development of local cultural nonprofit organizations, and the larger the size of the cultural nonprofit organizations in the region.

Based on the above analysis, the following hypotheses are proposed:

Hypothesis (H3).

The higher the dependency ratio of the elderly population is in a region, the larger the size of the cultural nonprofit organizations.

Hypothesis (H4).

The higher the illiteracy rates are in a region, the larger the size of the cultural nonprofit organizations.

Hypothesis (H5).

The higher the unemployment ratio is in a region, the larger the size of cultural nonprofit organizations.

2.2. Resource Supply

Demand theory explains only the necessary and insufficient conditions for the development of nonprofit organizations, whereas supply theory holds that the development of nonprofit organizations also requires an adequate supply of economic and human resources. Thus, the development of a regional cultural nonprofit organization depends on the supply of local economic resources and human resources. More specifically, the more abundant the resources are, the larger the size of nonprofit organizations.

First, the development of cultural nonprofit organizations requires adequate economic resources, with much of the nonprofit revenue coming from government funding [27]. Salamon et al. [2] estimated that among the 41 countries in their study, government support makes up approximately 35% of nonprofit sector revenue. By portraying nonprofits as complementary to the government, interdependence theory explains this phenomenon well [28]. Interdependence theory holds that nonprofit organizations and governments are cooperative partners in solving public problems for the following two reasons. On the one hand, there is a phenomenon of voluntary failure in the development of nonprofit organizations. To some extent, such organizations are faced with problems such as a lack of talent, a shortage of funds, and insufficient capacity. These problems hinder the development of nonprofit organizations, especially with the intensification of competition among nonprofit organizations, and due to the economic downturn, the shortage of funds has become a serious problem faced by a large number of nonprofit organizations [29]. On the other hand, with the rise of the new public management movement, governments have realized that nonprofit organizations can compensate for the government by responding to the diverse needs that are not being met because of the “government failure” problem and can be more effective in improving the quality of public services. To this end, the government has entrusted a large number of public services to professional nonprofit organizations. As the scale of the government’s purchase of public services continues to expand, the government requires an increasingly larger number of nonprofit organizations to undertake government services. In turn, financial assistance from government outsourcing services has become the main source of funding for these nonprofit organizations, which increasingly rely on government financial funds to maintain their operations. In this way, a mutual dependence is formed between the government and the nonprofit organizations [30]. In China, since the 16th National Congress of the Communist Party of the China in 2004, the fourth plenary session proposed “strong social construction and management [to] promote social management system innovation,”, the government has continued to promote China’s modern governance system, and as a consequence, the nonprofit organizations have gradually been incorporated into the framework of social governance and become partners with the Chinese government in areas of social governance and public services [31]. In recent years, the Chinese government’s purchase of services has become an important means of social governance innovation, with provinces issuing policies to purchase various types of public services from nonprofit organizations; increasing the purchase of old-age care services, public cultural services, and other areas, encouraging nonprofit organizations to enter the field of public services. In addition, the Chinese government has allocated special subsidy funds through the budget of the Ministry of Civil Affairs to support nonprofit organizations in participating in social services. For example, considering that the total budget for 2018 was approximately RMB 190 million, we believe that nonprofit organizations and governments in China are interdependent in providing public services. The more local governments fund public cultural services, the more they promote the development of local cultural nonprofit organizations. Thus, the following hypothesis is proposed:

Hypothesis (H6).

The greater the government funding of public cultural services is in a region, the greater the size of the cultural nonprofit organizations.

Social donations, service charges, and membership fees also constitute important sources of revenue for nonprofit organizations. The Johns Hopkins Comparative Nonprofit Sector Project estimated that, based on an average of 34 countries, 12% of the total revenues for nonprofit organizations came from charitable donations [2]. Social donations have become one of the important sources of income for nonprofit organizations. In recent years, the total amount of social donations in China has been increasing, from only RMB 1.4 billion in 1997 to more than RMB 75 billion in 2017; however, there are significant differences in the level of social donations in different regions of China. For example, the level of social donations in the eastern region is significantly higher than that in the central and western regions [32]. As the level of social donations in a region reflects, to some extent, the philanthropic culture in the region, regions with a stronger philanthropic culture are more attractive to nonprofit organizations. Research suggests that there is a positive relationship between the level of social donations and the size of the nonprofit organizations in a region [33,34]. For example, using data from the National Center for Charitable Statistics, Bielefeld [34] found that the level of gifts and grants to nonprofits was associated with larger nonprofit sectors. Although an analysis of the data from 34 countries revealed that service charges and membership fees are the largest source of revenue for nonprofits, accounting for approximately 53% of the total revenue [2], the revenue of nonprofit organizations from services and membership dues depends on the level of local economic development and the purchasing power of the residents. Empirical evidence supports the significant positive relationship between per capita income and the size of social service nonprofit organizations. Corbin [5], for example, found that U.S. metropolitan areas with higher income levels have larger nonprofit sectors in social services. This is because the purchasing power and economic level in the region provide resources for the establishment and operation of these organizations, and accordingly, the nonprofit organizations are more inclined to select sites in economically developed areas to obtain more economic resources. Moreover, an increase in people’s economic level compels them to demand more social services from nonprofit organizations. For example, Courty and Zhang [18] found strong support for the elitism hypothesis in China. More specifically, income increases participation in a broad range of cultural activities, including public cultural activities. In this way, economic level plays the role of both demand and supply in the development of regional nonprofit organizations. Based on the above analysis, we propose the following hypotheses:

Hypothesis (H7).

The higher the level of social donations is in a region, the larger the size of the cultural nonprofit organizations.

Hypothesis (H8).

The higher the economic level is in a region, the larger the size of the cultural nonprofit organizations.

In addition to economic resources, human resources are also necessary resources for the development of cultural nonprofit organizations. First, many organizations are established by their founders based on their own experiences, and thus, the vision of the founders is reflected in the process of organizational development. Entrepreneurship theory calls these founders social entrepreneurs [3]. Different from commercial entrepreneurs who create monetary value for companies, social entrepreneurs want to create and maintain social value (e.g., by providing public goods [35]). As these entrepreneurial activists are motivated by, and are able to create and operate nonprofit organizations, the development of regional nonprofit organizations requires the existence of these social entrepreneurs [36]. Social entrepreneurs are typically people of high socioeconomic status, such as those with high levels of education, as they are the ones who acknowledge the concepts of civil society, pay more attention to vulnerable groups and social problems, and have the consciousness and ability to help solve society’s problems. The research also shows that there exists a positive correlation between the proportion of a region’s population of those with higher levels of education and the size of the region’s nonprofit organizations. For example, Van Puyvelde and Brown [37] found that counties with populations with higher levels of education had larger nonprofit sectors in various fields, including the arts, culture, and humanities. Second, the proportion of people employed in cultural institutions reflects, to some extent, the level of cultural professionals in a region. The more cultural professionals there are in a region, the stronger the local cultural atmosphere will be and the greater the demand for cultural activities. Accordingly, these cultural professionals are more likely to found nonprofit organizations based on social needs, thus stimulating the development of local cultural nonprofit organizations. In addition, given the division of labor in the field of nonprofit organizations, the services that nonprofit organizations provide are becoming increasingly more professional [38]. Due to a lack of professional knowledge and greater mobility among volunteers, however, it is no longer possible to meet the needs of an organization by simply relying on volunteers, and thus, nonprofit organizations have an increasing demand for professional talent [20]. To increase the degree of professionalization in nonprofits, since 2006, China has officially included people engaged in social assistance programs, social charities, health services, youth services, rehabilitation for the disabled, and other social services in the category of professional technicians. Furthermore, social work has become a new profession. According to the Ministry of Civil Affairs, by the end of 2017, although the number of social work professionals in China was 1,025,757, there exists significant regional differences, as areas with abundant professional talent will attract nonprofit organizations, and the volunteer spirit and professional knowledge possessed by professionals promote the establishment of nonprofit organizations. Based on the above analysis, the following hypotheses are proposed:

Hypothesis (H9).

The greater the population with high levels of education is in a region, the larger the size of cultural nonprofit organizations.

Hypothesis (H10).

The more employees there are in cultural institutions in a region, the larger the size of cultural nonprofit organizations.

Hypothesis (H11).

The more social workers there are in a region, the larger the size of cultural nonprofit organizations.

2.3. Agglomeration Effect

Industrial agglomeration, which is an important economic and geographic phenomenon that occurs during the process of industrialization, provides multiple benefits to organizations [39]. Similarly, empirical studies have found that nonprofit organizations in counties, cities, and metropolitan areas display an agglomeration pattern beyond the impact of local needs and resource supplies [11,16,40,41,42]. For example, Katz [42] found that nonprofit human service organizations in Los Angeles were highly concentrated, but that their locations only partially overlapped with locations of high poverty. This is because these organizations only slowly developed an awareness of the benefits that agglomeration conveys to their organizations, and therefore, they chose to locate nearby to other similar organizations [42]. Marchesini da Costa [19], who analyzed nonprofit organizations in 5562 Brazilian municipalities, concluded that there is an agglomeration effect regarding the location of nonprofit organizations; however, neither access to resources nor poor socioeconomic indicators have powerful influences on their location. Rather, the main predictor of nonprofit entry is a high pre-existing density of nonprofits in that area [8].

Research indicates that labor market pooling, knowledge spillover, and cost reduction can explain this industrial agglomeration effect among nonprofit organizations [11,42,43]. As nonprofits need volunteers and professionals, a larger workforce can produce internal economies of scale when they provide services and benefit from higher quality job matches due to labor market pooling. Similarly, nonprofits can learn from other organizations about effective ways to raise money and reach their target groups, thus resulting in knowledge spillovers [43]. Partnering with other organizations to obtain customer demand information and product feasibility information can also reduce customer search costs, facilitate transactions and communications between upstream and downstream nonprofit organizations, and reduce transaction costs [42].

In addition, population ecology theory explains the formation process of clustering, as it focuses on the impact of the external environment on organizational development, emphasizes the importance of organizational density in the growth of new organizations, and holds that existing regional organizational density influences the growth of new organizations with respect to legitimacy [44]. An increase in the size of regional nonprofit organizations involves a legal process for the population of organizations, thus substantiating the value of organizational form, especially given that China’s nonprofit organizations have long faced a crisis of legitimacy in their development. As a result of the dual-control system of nonprofit organizations, it is difficult for many nonprofit organizations to register with civil affairs departments, and thus, they cannot obtain legal status (the dual-control system requires that most nonprofit organizations not only register with civil affairs departments, but they must also be affiliated with, and supervised by, a government agency in its functional area [28].). Accordingly, many organizations have been forced to abandon registration or seek other channels to conduct activities, such as registering businesses with the Industrial and Commercial Administration Bureau as a for-profit company, or being affiliated with nonprofit organizations that have legal status [45]. As a consequence, the dual-control system causes China’s nonprofit organizations to excessively depend on the government, which seriously affects the benign development of nonprofit organizations. For this reason, some regions have begun to explore reforms of nonprofit organization registration and management systems, although each region’s policy reform process is different; hence, many organizations choose to register in areas where the policy is less restrictive. With increases in the number of nonprofit organizations in a region, the legal status of nonprofit organizations in the region increases, which further stimulates the establishment of new organizations and the migration of existing organizations from other regions. Based on the above analysis, the following hypothesis is proposed:

Hypothesis (H12).

The growth of regional cultural nonprofit organizations is positively affected by the size of existing cultural nonprofit organizations in the region.

3. Research Methodology

3.1. Variables

In this study, the number of organizations is used to reflect the size of the cultural nonprofit organizations in a region, and the number of organizations added each year is used to reflect the growth in cultural nonprofit organizations in a region. In addition, this paper also studies the development of three types of cultural nonprofit organizations, including cultural social organizations, cultural private non-enterprise organizations, and cultural foundations. Since there is a strong positive correlation between the absolute number of cultural nonprofit organizations and the total regional population, to eliminate the influence of population, size, organizational density is used as the dependent variable. The density of cultural nonprofit organizations (NPOSIZE) is measured by the number of organizations per 100,000 people. As the total number of cultural foundations is small, the density of cultural foundations is measured by the number of foundations per million people [12]. The density of new organizations (NEWNPOSIZE) is measured by the number of new nonprofit organizations per million people.

The independent variables of this study are divided into three categories: regional demand factors, resource supply factors, and agglomeration effect factors. We use age heterogeneity (Age heterogeneity) [16] and education heterogeneity (Education heterogeneity) [17] to measure demand heterogeneity. In the Statistical Yearbook of China’s Population and Employment, age is divided into three groups: 0 to 14 years, 15 to 64 years, and 65 years and above. Education is divided into five levels: no schooling, primary school, middle school, high school, and junior college and above. In this paper, the Blau index is used to measure age heterogeneity and education heterogeneity, and the formula is: $1-{{\displaystyle \sum }}^{}{P}_i^2$ where $P_i$ is the proportion of a certain group. The dependency ratios of the elderly population (Elder, the ratio between the elderly population and the working-age population), the illiteracy rate (Illiteracy, the proportion of people who are illiterate in the population aged 15 and above) and the unemployment rate (Unemployment, registered urban unemployment rate) are used to measure vulnerable group demands [12,13]. Cultural services funding per capita (Government funding, regional cultural services funding/total population) is used to measure the government funding for public cultural services [9]. Resource supply factors include regional economic resources and human resources. Social donation per capita (Donation, social donation amount/total population) and regional economic level (Economic level, the per capita disposable income of urban residents) are used to measure a region’s economic resources [9]. Funding of regional cultural services, total social donations and per capita disposable income of urban residents are all given at the current year’s prices. To increase annual comparability, the original data are processed by using the consumer price index of each province as obtained from the Statistical Yearbook of China for 2009. In addition, the natural logarithm is used for cultural services funding per capita and the per capita disposable income of urban residents. Human resources are measured based on the population with higher education levels (HIGHEDUHigh education, the proportion of people with a college degree or above in the population aged 6 and above), employees in cultural institutions (Employee, the number of employees in cultural institutions per 10,000 people) and social workers (Social worker, the number of social workers and assistant social workers per 10,000 people) [12].

3.2. Data

Panel data for 31 provinces in mainland China from 2010 to 2015 are used as analysis samples. The data for variables were obtained directly from the Civil Affairs Statistics Yearbook of China (2011–2016), the Statistical Yearbook of China (2011–2016), the Statistical Yearbook of China’s Population and Employment (2011–2016), and the Cultural Relics Statistics Yearbook of China (2011–2016), or by calculation [46]. Age heterogeneity and education heterogeneity, elderly dependency ratio, illiteracy rate, unemployment rate, and the population with higher education levels were obtained from the Statistical Yearbook of China’s Population and Employment or were calculated according to the relevant data. Social donations per capita, the number of social workers and assistant social workers, the number of cultural nonprofit organizations, the number of cultural social organizations, the number of cultural private non-enterprise organizations, and the number of cultural foundations were obtained from the Civil Affairs Statistics Yearbook of China or calculated based on the relevant data. The per capita disposable incomes of urban residents were obtained from the Statistical Yearbook of China. Cultural services funding per capita and the number of employees in cultural institutions were calculated based on the Cultural Relics Statistics Yearbook of China. The descriptive statistics for each variable are presented in

Table 1. Descriptive statistics.

Variable Name	Mean	SD	Min	Max
Dependent variables
NPOSIZE overall	2.84	1.22	0.97	8.43
Social organizations NPOSIZE	2.00	0.80	0.70	4.98
Private non-enterprise organizations NPOSIZE	0.83	0.58	0	3.44
Foundations NPOSIZE	0.23	0.39	0	2.23
NEWNPOSIZE overall	2.46	3.28	−8.33	21.43
Social organizations NEWNPOSIZE	1.47	2.35	−8.97	11.21
Private non-enterprise organizations NEWNPOSIZE	0.96	1.80	−6.95	10.21
Foundations NEWNPOSIZE	0.03	0.15	−1.06	0.87
Independent variables
Age heterogeneity	0.41	0.04	0.28	0.49
Education heterogeneity	0.71	0.02	0.65	0.76
Elder (%)	12.36	2.62	6.71	20.04
Illiteracy (%)	6.15	6.17	1.46	41.18
Unemployment (%)	3.38	0.65	1.20	4.50
Government funding (ln)	3.51	0.59	2.25	5
Donation	24.30	36.22	0	190
Economic level (ln)	9.93	0.26	9.45	10.70
High education (%)	0.12	0.06	0.02	0.42
Employee (%)	6.13	2.58	2.90	25.81
Social worker (%)	0.89	1.51	0	11.09

.

3.3. Empirical Model

Before the regression analysis, the B-P test and Hausman test are conducted for each model to decide whether an OLS regression model, fixed-effects model, or a random-effects model is more appropriate. The results indicate that the model with size of cultural nonprofit organizations as dependent variables are suitable for the random-effects model, and the models with size of cultural social organizations and size of cultural private non-enterprise organizations as dependent variables are suitable for the fixed-effects model. The models with size of cultural foundations, size of new cultural nonprofit organizations, size of new cultural social organizations, size of new cultural private non-enterprise organizations, and size of new cultural foundations as dependent variables are suitable for the OLS regression model (

Table A1. Model Selection Results.

	Compare Type	Test Value	Compare Result	Selected Model
Model 1	OLS & (FE, RE)	χ2 (1) = 141.34, p = 0.000	(FE, RE) is superior to OLS	RE
	FE & RE	χ2 (10) = 3.46, p = 0.968	RE is superior to FE
Model 2	OLS & (FE, RE)	χ2 (1) = 185.53, p = 0.000	(FE, RE) is superior to OLS	FE
	FE & RE	χ2 (11) = 23.99, p = 0.013	FE is superior to RE
Model 3	OLS & (FE, RE)	χ2 (1) = 128.18, p = 0.000	(FE, RE) is superior to OLS	FE
	FE & RE	χ2 (11) = −1.97	FE is superior to RE
Model 4	OLS & (FE, RE)	χ2 (1) = 0.00, p = 1.000	OLS is superior to (FE, RE)	OLS
	FE & RE
Model 5	OLS & (FE, RE)	χ2 (1) = 0.03, p = 0.434	OLS is superior to (FE, RE)	OLS
	FE & RE
Model 6	OLS & (FE, RE)	χ2 (1) = 0.19, p = 0.333	OLS is superior to (FE, RE)	OLS
	FE & RE
Model 7	OLS & (FE, RE)	χ2 (1) = 0.00, p = 1.000	OLS is superior to (FE, RE)	OLS
	FE & RE
Model 8	OLS & (FE, RE)	χ2 (1) = 0.00, p = 1.000	OLS is superior to (FE, RE)	OLS
	FE & RE

Notes: FE denotes fix-effect model, RE denotes random-effect model.

).

4. Results and Discussion

The regression results are presented in

Table 2. Regression results (Model 1 to Model 4).

Variable	Model 1	Model 2	Model 3	Model 4
	Nonprofit Organizations Overall	Social Organizations	Private Non-Enterprise Organizations	Foundations
Age heterogeneity	6.978 * (3.671)	4.395 (3.758)	5.549 * (2.883)	−10.945 (10.860)
Education heterogeneity	16.423 *** (4.212)	9.442 *** (3.342)	0.180 (2.564)	−14.395 (15.443)
Elder	−0.032 (0.054)	−0.027 (0.049)	−0.019 (0.038)	0.016 (0.155)
Illiteracy	0.003 (0.027)	0.046 * (0.027)	0.020 (0.021)	0.006 (0.075)
Unemployment	−0.223 (0.161)	−0.186 (0.127)	−0.124 (0.098)	−1.154 ** (0.519)
Government funding	0.582 ** (0.281)	−0.145 (0.258)	0.556 *** (0.198)	0.433 (0.933)
Donation	0.001 (0.002)	−0.001 (0.002)	0.001 (0.001)	0.009 (0.012)
Economic level	1.381 ** (0.567)	1.399 *** (0.468)	−0.031 (0.359)	1.180 (1.845)
High education	−4.562 * (2.687)	−1.178 (2.223)	1.823 (1.708)	−0.757 (9.744)
Employee	−0.086 ** (0.038)	−0.042 (0.026)	−0.075 *** (0.020)	−0.190 (0.179)
Social worker	0.093 * (0.056)	0.038 (0.036)	0.080 *** (0.028)	0.052 (0.328)
Constant	−25.407 (5.670)	−18.868 (4.678)	−2.525 (3.589)	7.023 (19.752)
Within R2	0.4899	0.4664	0.4475
Between R2	0.2620	0.0343	0.0267
Overall R2	0.3183	0.1005	0.0777	0.1289
F		11.44 ***	10.60 ***	2.34 **
Wald χ2	144.49 ***			25.75 ***

Notes: Values in parentheses denote standard deviations; *** p < 0.01; ** p < 0.05; * p < 0.1.

and

Table 3. Regression results (Model 5 to Model 8).

Variable	Model 5	Model 6	Model 7	Model 8
	New Nonprofit Organizations Overall	New Social Organizations	New Non-Enterprise Organizations	New Foundations
Age heterogeneity	−3.273 (8.654)	−0.670 (6.691)	−2.680 (4.721)	0.011 (0.345)
Education heterogeneity	11.038 (14.956)	12.468 (10.878)	4.791 (7.956)	−2.871 *** (0.520)
Elder	0.057 (0.123)	−0.033 (0.095)	0.082 (0.068)	−0.001 (0.005)
Illiteracy	0.004 (0.061)	−0.024 (0.047)	0.015 (0.033)	0.011 *** (0.003)
Unemployment	−0.611 (0.424)	−0.307 (0.328)	−0.318 (0.227)	0.064 *** (0.016)
Government funding	−0.628 (0.760)	−0.760 (0.592)	0.242 (0.406)	0.069 ** (0.030)
Donation	0.085 * (0.010)	0.006 (0.007)	0.013 ** (0.005)	0.001 * (0.0004)
Economic level	4.181 *** (1.474)	2.377 ** (1.137)	1.845 ** (0.806)	−0.044 (0.059)
High education	−16.545 * (8.574)	−11.583 * (6.735)	−8.392 * (4.360)	1.949 *** (0.337)
Employee	−0.020 (0.142)	0.059 (0.110)	−0.074 (0.078)	−0.006 (0.006)
Social worker	0.001 (0.261)	0.108 (0.202)	−0.074 (0.142)	0.002 (0.011)
NPOSIZE overall last year	0.740 *** (0.286)
Social organizations NPOSIZE last year		0.482 (0.311)
Private non-enterprise organizations NPOSIZE last year			0.700 ** (0.325)
Foundations NPOSIZE last year				−0.351 *** (0.048)
Constant	−42.353 (16.89)	−26.613 (12.674)	−19.878 (9.226)	2.261 (0.647)
Overall R2	0.2483	0.1197	0.2584	0.4062
F	4.76 ***	1.96 **	5.02 ***	9.86 ***

Notes: Values in parentheses denote standard deviations; *** p < 0.01; ** p < 0.05; * p < 0.1.

.

The dependent variables in Model 1 to Model 4 are the sizes of the cultural nonprofit organizations, sizes of the cultural social organizations, sizes of the cultural private non-enterprise organizations, and sizes of the cultural foundations. As presented in Table 2, the explanatory power of the four models for the overall difference in the sizes of the nonprofit organizations is 31.83%, 10.05%, 7.77%, and 12.89%, respectively. Both regional demand and resource supply have effects on the size of the cultural nonprofit organizations, but there are also differences in the effects on the three types of nonprofit organizations.

First, we investigated the effect of regional demand on the size of the nonprofit sector. The results reveal that age heterogeneity and education heterogeneity have significant positive relationships with the size of cultural nonprofit organizations, which indicates that the models support Hypothesis 1 and Hypothesis 2. Consistent with previous research findings [5,9,15], the results confirm the arguments of the government failure theory (i.e., an increase in demand heterogeneity for public services has positive effects on the size of the nonprofit sector [1]). Further analysis reveals that demand heterogeneity has different effects on the size of the different types of nonprofit organizations. Education heterogeneity has a significant positive relationship with the size of cultural social organizations, and age heterogeneity has a significant positive relationship with the size of cultural private non-enterprise organizations; however, there is no significant relationship between the size of cultural foundations and demand heterogeneity.

Second, we argued that the needs of vulnerable groups have a positive influence on the development of cultural nonprofit organizations. As the parameter estimate on the dependency ratio of the elderly population is insignificant, the findings do not support Hypothesis 3, and Hypothesis 4 is only partially supported. More specifically, the illiteracy rate has a significant positive influence on the size of cultural social organizations. In contrast to our expectations, the unemployment rate has a significant negative influence on the size of cultural foundations; however, this finding supports the results of Saxton and Benson [4], and Joassart-Marcelli and Wolch [13].

Third, we investigated the effect of economic resources on the size of the cultural nonprofit sector. Hypothesis 6 holds that government funding for cultural services promotes the development of cultural nonprofit organizations, and the results in Table 2 support this hypothesis. Consistent with the results of Lecy and Van Slyke [47], Bae and Sohn [48], and Kim [49], there is a significant positive relationship between government cultural funding and the size of cultural nonprofit organizations. This result indicates that Chinese government funding for public culture has not impeded the development of nonprofit organizations. Rather, in contrast, it has stimulated the development of local cultural nonprofit organizations. According to the interdependency theory, an increase in public funding leads to the expansion of the size of nonprofit sectors [29]. This result supports the interdependency theory that government and nonprofit organizations have become partners in providing public cultural services in China. Concerning the level of social donation and economic level, we found that the parameter estimate for social donations is insignificant, whereas the parameter estimate for economic levels is significant. This finding is consistent with the results of Van Puyvelde and Brown [37] and Lecy and Van Slyke [47], which show that levels of income have a positive effect on the size of the nonprofit sector in the field of human services; hence, we find support for and Hypothesis 8 but not for Hypothesis 7.

Fourth, we predicted that human resources in a region are positively associated with the size of the cultural nonprofit sector; however, we found mixed results. We assumed that higher education levels imply more available human resources, and consequently more social entrepreneurs to establish cultural nonprofit organizations. Surprisingly, the level of education is negatively associated with the size of the cultural nonprofit sector, and thus, the results do not support Hypothesis 9. This is contrary to the findings of Van Puyvelde and Brown [37], Ben-Ner and Hoomissen [17], and Grønbjerg and Paarlberg [14], all of whom concluded that a region with a higher level of education has a greater nonprofit sector. This result may be related to the lack of a strong social entrepreneurship atmosphere in China from 2010 to 2015. Imperfect policy and incentive measures have led to the failure of professionals with high education levels to enter the field of nonprofit organization. According to a report, in 2010, 24% of Chinese charitable foundations did not have any full-time staff [50]. Hypothesis 10 expects that higher proportions of employees in regional cultural institutions will enhance the size of local cultural nonprofit organizations; however, the results indicate the opposite (i.e., the higher the proportion of employees in cultural institutions is, the smaller the size of regional cultural nonprofit organizations). Thus, Hypothesis 10 is not supported. This may be because the proportion of employees in cultural institutions reflects the scale of public cultural services directly provided by a regional government’s public cultural institutions. Different from the impact of government funding on the development of cultural nonprofit organizations, the scale of the government’s direct provision of public cultural services reduces people’s demand for cultural nonprofit organizations, thus compressing the development space available for the nonprofit sector. This finding is consistent with those of Matsunaga and Yamauchi [51], and Van Puyvelde and Brown [37], and provides evidence for government failure theory (i.e., an increase in government direct expenditure has a negative effect on the size of the nonprofit sector [5]). With respect to the proportion of social workers, we found a significant positive impact on the size of cultural nonprofit organizations. Thus, Hypothesis 11 is verified; however, there are different effects on different types of nonprofit organizations. Only the influence on the size of cultural private non-enterprise organizations is statistically significant, which may be related to the greater demand for social workers from cultural private non-enterprise organizations as the main provider of nonprofit public services in China.

The dependent variables of Model 5 to Model 8 are the size of new cultural nonprofit organizations, size of new cultural social organizations, size of new cultural private non-enterprise organizations and size of new cultural foundations. As presented in Table 3, the explanatory power of the four models for the overall differences in the size of nonprofit organizations are 24.83%, 11.97%, 25.84%, and 40.62%, respectively. These four models analyze the impact of the number of existing cultural nonprofit organizations on the growth of new nonprofit organizations in the region, to explore whether there is an agglomeration effect in the development of regional cultural nonprofit organizations. The results indicate that the size of cultural nonprofit organizations in the prior year significantly increases the size of new cultural nonprofit organizations in the next year; therefore, Hypothesis 12 is verified. This result is consistent with previous findings with respect to the agglomeration effect at the metropolitan level [11], city level [13], and county level [16,42], and adds new evidence to the agglomeration effect theory in the nonprofit sector. Under the influence of the agglomeration effect, China’s cultural nonprofit organizations are unevenly distributed and present an agglomeration development model that may be detrimental to the goal of equalizing public services [13]. For this reason, on the one hand, the Chinese government must continue to promote the reformation of local registration and management systems, give more power to local authorities, and provide policy support for nonprofit organizations. In addition, the government should support the development of cultural nonprofit organizations in disadvantaged areas through initiatives such as nonprofit organization incubators, social innovations, and venture philanthropy. On the other hand, the government should continue to optimize the supply structure for public culture by streamlining public cultural institutions and promoting the transformation of institutions into nonprofit organizations to create space for the development of cultural nonprofit organizations and capitalize on the advantages of nonprofit organizations in providing cultural services. Further analysis indicates that the agglomeration effect differs for the three types of nonprofit organizations, and there is an agglomeration effect in the growth of cultural private non-enterprise organizations. In contrast, there is a restraining effect rather than agglomeration effect in the growth of cultural foundations. Furthermore, the size of cultural foundations in the prior year significantly inhibits the growth of foundations in the next year. This finding suggests that the influence of local nonprofit organization ecology on foundations differs from that of private non-enterprise organizations. This may be related to China’s Regulations on the Management of Foundations, which requires foundations to be registered and managed by the state and provincial civil affairs departments (since August 2012, the Ministry of Civil Affairs of China has gradually delegated the authority for the registration and approval of non-public fundraising foundations), whereas private non-enterprise organizations may be registered and managed by the state, and civil affairs departments at or above the county level; therefore, compared with private non-enterprise organizations, foundations face more stringent audits with respect to registration and management, which impedes development; therefore, the size of previous cultural foundations in a region suppresses the establishment of new foundations.

5. Conclusions

Using panel data from the 31 provinces in mainland China from 2010 to 2015, this study examines the factors influencing the development and sustainable growth of Chinese cultural nonprofits. The results reveal that the development of cultural nonprofit organizations responds to the demand for public cultural services in a region. More specifically, age heterogeneity and education heterogeneity are positively associated with the size of cultural nonprofit organizations, and the illiteracy rate has a significant positive influence on the size of cultural social organizations. We also find that regional economic resources and human resources have positive impacts on the size of cultural nonprofit organizations. More specifically, government funding for cultural services, the economic level, and the proportion of social workers are positively associated with the size of cultural nonprofit organizations. Finally, we find that there is an agglomeration effect in the development of regional cultural nonprofit organizations.

This study makes two theoretical contributions. First, this study developed a theoretical framework for understanding the regional development of the cultural nonprofit sector in China from the perspectives of regional demand, resource supply, and agglomeration effects. Second, it extends the research on the growth of nonprofit organizations to non-Western countries; therefore, this study enhances the geographical and cultural diversity of nonprofit organizations research and enriches the literature by contributing empirical evidence from China.

These findings have important practical implications for the government to promote the development of cultural nonprofits. Significant differences in the regional development of Chinese cultural nonprofits are not only determined by regional demand, but also significantly related to the regional supply and ecological environments; therefore, on the one hand, policy makers should promote the development of regional cultural nonprofits by enhancing the government purchase of services and cultivating professionals in social work. On the other hand, the Chinese government must promote reforms in the registration and management systems of nonprofits and support the development of cultural nonprofits in disadvantaged areas through various preferential initiatives.

This study has some research limitations and topics that require further attention. First, the study used the number of nonprofit organizations in the current year minus the number of nonprofit organizations in the previous year to represent the number of new nonprofit organizations in the current year. This value is not the exact number of newly established nonprofits, because some nonprofits are deregistered or banned every year; however, we do not think this discrepancy has much impact on the conclusions of the study, because the number of nonprofits deregistered or banned is minimal. Second, we were able to obtain data on regional cultural nonprofit organizations from 2007 to 2015 from the Statistical Yearbook of Civil Affairs of China; however, because data such as social donation data from 2007 to 2009 cannot be obtained, data analysis of this information could not be performed. Therefore, in this study, cultural nonprofit organizations in various provinces from 2010 to 2015 were selected as samples. As China’s nonprofit sector has undergone various stages of development, there may be differences in the influencing mechanisms of the development of nonprofit sectors during the different stages. More meaningful conclusions may be obtained by collecting study samples for longer time periods and analyzing them based on the developmental stage. Third, to ensure statistical significance, the number of data samples can be increased by subdividing these provinces into lower levels in the future.